{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "## Running training in 2019 and 2020: time series\n", "\n", "> [https://github.com/BMClab/covid19](https://github.com/BMClab/covid19) \n", "> [Laboratory of Biomechanics and Motor Control](http://pesquisa.ufabc.edu.br/bmclab/) \n", "> Federal University of ABC, Brazil\n", "\n", "**The data used in this Jupyter notebook are available on the Figshare repository https://doi.org/10.6084/m9.figshare.16620238.**" ] }, { "cell_type": "markdown", "metadata": { "toc": true }, "source": [ "

Contents

\n", "
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Setup" ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:01.062214Z", "start_time": "2022-03-12T03:51:00.291390Z" } }, "outputs": [], "source": [ "from pathlib import Path\n", "import pickle\n", "import numpy as np\n", "import scipy as sp\n", "import pandas as pd\n", "import seaborn as sns\n", "import matplotlib\n", "import matplotlib.pyplot as plt\n", "%matplotlib inline\n", "import matplotlib.patches as patches\n", "from tqdm.notebook import tqdm\n", "%load_ext watermark " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Environment" ] }, { "cell_type": "code", "execution_count": 45, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T04:10:26.224225Z", "start_time": "2022-03-12T04:10:26.158236Z" } }, "outputs": [ { "data": { "text/html": [ "" ], "text/plain": [ "[(0.0, 0.4196078431372549, 0.6431372549019608),\n", " (1.0, 0.5019607843137255, 0.054901960784313725),\n", " (0.6705882352941176, 0.6705882352941176, 0.6705882352941176),\n", " (0.34901960784313724, 0.34901960784313724, 0.34901960784313724),\n", " (0.37254901960784315, 0.6196078431372549, 0.8196078431372549),\n", " (0.7843137254901961, 0.3215686274509804, 0.0),\n", " (0.5372549019607843, 0.5372549019607843, 0.5372549019607843),\n", " (0.6352941176470588, 0.7843137254901961, 0.9254901960784314),\n", " (1.0, 0.7372549019607844, 0.4745098039215686),\n", " (0.8117647058823529, 0.8117647058823529, 0.8117647058823529)]" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Last updated: 2022-03-12T01:10:26.186979-03:00\n", "\n", "Python implementation: CPython\n", "Python version : 3.9.10\n", "IPython version : 7.32.0\n", "\n", "Compiler : GCC 9.4.0\n", "OS : Linux\n", "Release : 5.13.0-30-generic\n", "Machine : x86_64\n", "Processor : x86_64\n", "CPU cores : 12\n", "Architecture: 64bit\n", "\n", "pandas : 1.4.1\n", "numpy : 1.22.3\n", "scipy : 1.8.0\n", "seaborn : 0.11.2\n", "matplotlib : 3.5.1\n", "bootstrap_stat: 0.2.4.2\n", "json : 2.0.9\n", "\n" ] } ], "source": [ "path2 = Path(r'./../data/')\n", "\n", "\n", "country = ['Brazil', 'br']\n", "country = ['all countries', 'all']\n", "\n", "\n", "load_results = True # load file with results if exists\n", "if not load_results:\n", " # Warning: the bootstrap process takes hours!\n", " from bootstrap_stat import bootstrap_stat as bp\n", "\n", "tqdm.pandas(desc='Boot', leave=False)\n", "pd.set_option('display.precision', 3)\n", "plt.rcParams.update({'font.size': 14, 'xtick.labelsize': 12,\n", " 'ytick.labelsize': 12})\n", "sns.set_style('whitegrid', rc={'xtick.bottom': True, 'xtick.top': True,\n", " 'ytick.left': True,\n", " 'ytick.right': True, 'xtick.direction': 'in',\n", " 'ytick.direction': 'in'})\n", "Slice = pd.IndexSlice\n", "\n", "# Set of colors that is unambiguous both to colorblinds and non-colorblinds\n", "# https://www.cta-observatory.org/wp-content/uploads/2020/10/CTA_ColourBlindness_BestPractices-1.pdf\n", "# https://jfly.uni-koeln.de/color/index.html\n", "#matplotlib.style.use('seaborn-colorblind')\n", "matplotlib.style.use('tableau-colorblind10')\n", "colors = sns.color_palette()\n", "display(colors)\n", "\n", "colorbase = 'k'\n", "color2019 = colors[0]\n", "color2020 = colors[1]\n", "colorn = colors[5]\n", "colorvar = colors[4]\n", "colorind = colors[6]\n", "\n", "# Statistics\n", "alpha = 0.05\n", "two_tailed = True\n", "# confidence interval\n", "estimate = np.mean\n", "boot = 100000 # number of bootstrap samples\n", "seed = 0\n", "\n", "# other variables:\n", "years = ['2019', '2020']\n", "quarters = ['First', 'Second', 'Third', 'Fourth']\n", "months = ['January', 'February', 'March', 'April', 'May', 'June', 'July',\n", " 'August', 'September', 'October', 'November', 'December']\n", "freqs = ['d', 'w', 'm', 'q']\n", "freqs_s = ['day', 'week', 'month', 'quarter']\n", "# feature (dependent variable)\n", "variables, units = ['distance', 'duration'], ['km', 'min']\n", "vars_d = [v + '_d' for v in variables]\n", "# period for analysis\n", "freq, freq_s = freqs[1], freqs_s[1]\n", "\n", "%watermark\n", "%watermark --iversions" ] }, { "cell_type": "markdown", "metadata": { "code_folding": [ 6, 12, 18, 82, 96, 101, 106, 111, 123 ] }, "source": [ "### Helping functions" ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:01.109888Z", "start_time": "2022-03-12T03:51:01.090196Z" }, "code_folding": [ 4, 10, 16, 53, 82, 96, 101, 106, 111, 119, 123 ] }, "outputs": [], "source": [ "# standardization (mean 0, variance 1)\n", "stdz = lambda x: (x-x.mean())/x.std()\n", "\n", "\n", "def normality(x):\n", " \"\"\"Kurtosis, skewness, p-value of normality test using the scipy library.\n", " \"\"\"\n", " return [sp.stats.kurtosis(x), sp.stats.skew(x), sp.stats.normaltest(x).pvalue]\n", "\n", "\n", "def cohensd(x):\n", " \"\"\"Cohen's d effect size for within-subject sample.\n", " \"\"\"\n", " return np.abs(np.mean(x, axis=0)) / np.std(x, axis=0, ddof=1)\n", "\n", "\n", "def ci_asl(y, stat=np.nanmean, alpha2=0.025, boot=boot):\n", " \"\"\"Confidence interval and 'pvalue' of hypothesis testing by bootstrap.\n", "\n", " Parameters\n", " ----------\n", " y : 1-d array_like\n", " Calculate the confidence interval of these values.\n", " stat : function (optional, default=np.nanmean)\n", " Statistics (a function) to which the confidence interval will be\n", " estimated\n", " alpha : float (optional, default=0.025)\n", " alpha level (]0, 1[) of the confidence interval\n", " boot : int (optional, default=100000)\n", " Number of bootstrap iteractions\n", "\n", " Returns\n", " -------\n", " [ci_low, ci_high, asl] : list of floats\n", " Lower and upper bounds on a 100(1-2*`alpha`)% confidence interval on\n", " stat of `y`.\n", " Achieved significance level, the probability of an outcome at least as\n", " extreme as that actually observed under the null hypothesis; the p-value.\n", "\n", "\n", " See Also\n", " --------\n", " https://github.com/rwilson4/bootstrap-stat\n", " \"\"\"\n", " ed = bp.EmpiricalDistribution(y)\n", " ci_low, ci_high, theta_star = bp.percentile_interval(ed, stat, alpha=alpha2, B=boot,\n", " return_samples=True, num_threads=-1)\n", " asl = bp.percentile_asl(ed, stat, y, theta_0=0, B=boot,\n", " theta_star=theta_star, two_sided=True, num_threads=1)\n", " \n", " return [ci_low, ci_high, asl]\n", "\n", "\n", "def describe(y, variable=['distance_d', 'duration_d'],\n", " new_variable=None, show=True):\n", " \"\"\"Descriptive statistics for dataframe columns.\n", " \"\"\"\n", " ys = []\n", " stat = ['count', 'mean', 'std', cohensd]\n", " for var in variable:\n", " y1 = y[['athlete', var]\n", " ].pivot(columns='athlete')[var].T.agg(stat, axis=0).transpose()\n", " y2 = y[['athlete', var]\n", " ].pivot(columns='athlete')[var].T.progress_apply(ci_asl, alpha2=alpha/2\n", " ).transpose()\n", " y2.rename(columns={0: 'ci_inf', 1: 'ci_sup', 2: 'pvalue'}, inplace=True)\n", " ys.append(pd.concat([y1, y2], axis=1))\n", " if new_variable is None:\n", " new_variable = variable\n", " ys = pd.concat(ys, axis=1, keys=new_variable)\n", " ys.index = ys.index.dayofyear # range(ys.shape[0])\n", " # correct index when joining leap year and non leap year\n", " if ys.shape[0] == 365: \n", " ys.index = range(1, 366, 1)\n", " elif ys.shape[0] == 52:\n", " ys.index = range(1, 365, 7)\n", " ys.index.rename('Period', inplace=True)\n", " if show:\n", " display(ys)\n", " return ys\n", "\n", "\n", "def highlight_cohensd(x):\n", " '''\n", " Set background color for cohensd values.\n", " '''\n", " #['Nothing', 'Very small', 'Small', 'Medium', 'Large', 'Very large', 'Huge']\n", " d = [0, .01, .2, .5, .8, 1.2, 2, 100]\n", " # colors\n", " cm = sns.color_palette('Greens', n_colors=len(d)-2).as_hex()\n", " cm = ['background-color: {};font-weight:bold;'.format(c) for c in cm]\n", " cm.insert(0, 'background-color: '';') \n", " x = pd.cut(x, bins=d, right=False, labels=cm, retbins=False)\n", " return x.to_list()\n", "\n", "\n", "def highlight_pvalue(x, level=alpha,\n", " props='color:white;background-color:red;font-weight:bold'):\n", " return np.where(x <= level, props, '')\n", "\n", "\n", "def highlight_max(x,\n", " props='color:white;background-color:blue;font-weight:bold'):\n", " return np.where(x == np.nanmax(x.values), props, '')\n", "\n", "\n", "def highlight_min(x,\n", " props='color:white;background-color:purple;font-weight:bold'):\n", " return np.where(x == np.nanmin(x.values), props, '')\n", "\n", "\n", "def display_df(df, subset='mean'):\n", " \"\"\"Diplay dataframe with a certain style.\n", " \"\"\"\n", " if type(df.index) == pd.DatetimeIndex:\n", " df.index = df.index.strftime('%Y-%m-%d')\n", " display(df.style\n", " .highlight_max(subset=subset,\n", " props='color:white;font-weight:bold;background-color:blue;')\n", " .highlight_min(subset=subset,\n", " props='color:white;font-weight:bold;background-color:purple;'))\n", " \n", "\n", "def display_dfsd(df, title=''):\n", " \"\"\"Display dataframe with certain style.\n", " See https://pandas.pydata.org/pandas-docs/stable/user_guide/style.html\n", " \"\"\"\n", " Slice = pd.IndexSlice\n", " display(df.style.format('{:.0f}', subset=Slice[:, Slice[:, 'count']])\n", " .format('{:.4f}', subset=Slice[:, Slice[:, 'pvalue']])\n", " .highlight_max(subset=Slice[:, Slice[:, 'mean']],\n", " props='color:white;font-weight:bold;background-color:blue;')\n", " .highlight_min(subset=Slice[:, Slice[:, 'mean']],\n", " props='color:white;font-weight:bold;background-color:purple;')\n", " .apply(highlight_cohensd, subset=Slice[:, Slice[:, 'cohensd']])\n", " .apply(highlight_pvalue, level=alphaS, subset=Slice[:, Slice[:, 'pvalue']])\n", " .set_caption(title)\n", " .set_table_styles([{'selector': 'caption', 'props': 'font-weight:bold; ' +\n", " 'text-align: center; ' +\n", " 'caption-side: top; color: #000000; font-size:1.4em;'\n", " }]))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Load dataset\n", "\n", "### Running activity\n", "\n", "Let's load the dataset and create a dictionary with all the data and grouped by gender and age. \n", "We will also create two equivalents dictionaries with the number of athletes per period and the number of activities per athlete." ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:04.304442Z", "start_time": "2022-03-12T03:51:01.111454Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Country: all countries\n" ] }, { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Year: 2019, Frequency: day (first and last rows)
 athletedistancedurationcountry
datetime    
2019-01-01 00:00:0000.0000000.000000United States
2019-12-31 00:00:003759825.670000223.000000China
\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Year: 2019, Frequency: week (first and last rows)
 athletedistancedurationcountry
datetime    
2019-01-01 00:00:0000.0000000.000000United States
2019-12-24 00:00:003759846.077500322.947917China
\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Year: 2019, Frequency: month (first and last rows)
 athletedistancedurationcountry
datetime    
2019-01-31 00:00:0000.0000000.000000United States
2019-12-31 00:00:0037598297.7451612378.951613China
\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Year: 2019, Frequency: quarter (first and last rows)
 athletedistancedurationcountry
datetime    
2019-03-31 00:00:0000.0000000.000000United States
2019-12-31 00:00:00375981206.5900008937.600000China
\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Year: 2020, Frequency: day (first and last rows)
 athletedistancedurationcountry
datetime    
2020-01-01 00:00:0000.0000000.000000United States
2020-12-31 00:00:003759810.62000071.000000China
\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Year: 2020, Frequency: week (first and last rows)
 athletedistancedurationcountry
datetime    
2020-01-01 00:00:0000.0000000.000000United States
2020-12-23 00:00:0037598102.052222847.220370China
\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Year: 2020, Frequency: month (first and last rows)
 athletedistancedurationcountry
datetime    
2020-01-31 00:00:0000.0000000.000000United States
2020-12-31 00:00:0037598377.7193552842.564516China
\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Year: 2020, Frequency: quarter (first and last rows)
 athletedistancedurationcountry
datetime    
2020-03-31 00:00:00034.040000180.166667United States
2020-12-31 00:00:00375981096.7600007969.533333China
\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "dfs, dfs_no0, nathletes, nruns = dict(), dict(), dict(), dict()\n", "print('Country: {}'. format(country[0]))\n", "for year in years:\n", " for f, frq in enumerate(freqs):\n", " # load dataset and select country from start to speedup process\n", " fname = path2 / 'run_ww_{}_{}.parquet'.format(year, frq)\n", " df = pd.read_parquet(fname, columns=['datetime', 'athlete', 'distance',\n", " 'duration', 'country'])\n", " if country[1] != 'all':\n", " df = df[df['country'] == country[0]]\n", " df = df.set_index('datetime', drop=True)\n", " #df.index = df.index.normalize()\n", " # correct dates\n", " if df.index[0].is_leap_year and frq == 'd':\n", " # if leap year, remove Feb 29, average data of Feb 28 and 29\n", " feb28 = pd.Timestamp(unit='D', year=df.index[0].year, month=2, day=28)\n", " feb29 = pd.Timestamp(unit='D', year=df.index[0].year, month=2, day=29)\n", " nbs = df.select_dtypes(include='number').columns.to_list()\n", " df.loc[df.index==feb28, nbs] = (df.loc[df.index==feb28, nbs].values +\n", " df.loc[df.index==feb29, nbs].values)/2\n", " df = df.drop(index=feb29) \n", " \n", " display(df.iloc[[0, -1]].style.set_caption(\n", " 'Year: {}, Frequency: {} (first and last rows)'.format(year, freqs_s[f])))\n", " # create dictionary will all the data (including no runs data)\n", " dfs[year, frq] = df\n", " # create dictionary will all the data (excluding no runs data)\n", " dfs_no0[year, frq] = df[df[variables[0]] > 0]\n", " # number of athletes and running activities\n", " nathletes[year, frq] = dfs_no0[year, frq].groupby(level='datetime').size()\n", " x = dfs_no0[year, frq].groupby('athlete').size()\n", " nruns[year, frq] = x[x > 0] # removes athletes with zero runs" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Basic information about data in the dataset" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:04.475734Z", "start_time": "2022-03-12T03:51:04.305491Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "DatetimeIndex: 13290380 entries, 2019-01-01 to 2019-12-31\n", "Data columns (total 4 columns):\n", " # Column Dtype \n", "--- ------ ----- \n", " 0 athlete category\n", " 1 distance float64 \n", " 2 duration float64 \n", " 3 country category\n", "dtypes: category(2), float64(2)\n", "memory usage: 381.5 MB\n" ] } ], "source": [ "dfs['2019', 'd'].info(memory_usage='deep')" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:04.485634Z", "start_time": "2022-03-12T03:51:04.476690Z" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
athletedistancedurationcountry
datetime
2019-01-0100.0000.000United States
2019-01-0115.27030.200Germany
2019-01-0129.30098.000United Kingdom
2019-01-013103.130453.400United Kingdom
2019-01-01434.670185.650United States
...............
2019-12-243759412.83563.481United Kingdom
2019-12-2437595151.725623.919United States
2019-12-243759635.770153.081United States
2019-12-243759731.938176.473United States
2019-12-243759846.078322.948China
\n", "

1893424 rows × 4 columns

\n", "
" ], "text/plain": [ " athlete distance duration country\n", "datetime \n", "2019-01-01 0 0.000 0.000 United States\n", "2019-01-01 1 5.270 30.200 Germany\n", "2019-01-01 2 9.300 98.000 United Kingdom\n", "2019-01-01 3 103.130 453.400 United Kingdom\n", "2019-01-01 4 34.670 185.650 United States\n", "... ... ... ... ...\n", "2019-12-24 37594 12.835 63.481 United Kingdom\n", "2019-12-24 37595 151.725 623.919 United States\n", "2019-12-24 37596 35.770 153.081 United States\n", "2019-12-24 37597 31.938 176.473 United States\n", "2019-12-24 37598 46.078 322.948 China\n", "\n", "[1893424 rows x 4 columns]" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "dfs['2019', freq]" ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:04.493669Z", "start_time": "2022-03-12T03:51:04.486511Z" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
athletedistancedurationcountry
datetime
2020-01-0100.0000.000United States
2020-01-01170.330394.200Germany
2020-01-01214.65079.067United Kingdom
2020-01-01341.410195.667United Kingdom
2020-01-01441.340209.100United States
...............
2020-12-2337594128.154572.587United Kingdom
2020-12-233759520.05187.461United States
2020-12-2337596144.636625.774United States
2020-12-23375970.0000.000United States
2020-12-2337598102.052847.220China
\n", "

1893424 rows × 4 columns

\n", "
" ], "text/plain": [ " athlete distance duration country\n", "datetime \n", "2020-01-01 0 0.000 0.000 United States\n", "2020-01-01 1 70.330 394.200 Germany\n", "2020-01-01 2 14.650 79.067 United Kingdom\n", "2020-01-01 3 41.410 195.667 United Kingdom\n", "2020-01-01 4 41.340 209.100 United States\n", "... ... ... ... ...\n", "2020-12-23 37594 128.154 572.587 United Kingdom\n", "2020-12-23 37595 20.051 87.461 United States\n", "2020-12-23 37596 144.636 625.774 United States\n", "2020-12-23 37597 0.000 0.000 United States\n", "2020-12-23 37598 102.052 847.220 China\n", "\n", "[1893424 rows x 4 columns]" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "dfs['2020', freq]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Load dataset with indexes of policy responses to the coronavirus pandemic" ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:04.499914Z", "start_time": "2022-03-12T03:51:04.494518Z" } }, "outputs": [], "source": [ "if country[1] == 'all':\n", " # Covid-19 stringency index averaged by percentage of athletes' countries\n", " c19idx = pd.read_csv(path2 / 'policy_response_indexes.csv',\n", " sep=',', header=0, parse_dates=['date'], verbose=False)\n", "else:\n", " # Covid-19 stringency index of each country\n", " c19idx = pd.read_csv(path2 / 'policy_response_indexes_countries.csv',\n", " sep=',', header=0, parse_dates=['date'], verbose=False)\n", " c19idx = c19idx[c19idx['country'] == country[0]].drop(columns='country')\n", " \n", "c19idx = c19idx.set_index('date', drop=True)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Correct dates" ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:04.518196Z", "start_time": "2022-03-12T03:51:04.501000Z" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
sicihiwi
date
2020-01-010.1780.3590.0000.000
2020-01-080.2730.4480.0000.000
2020-01-150.7221.3480.0000.000
2020-01-222.9065.6260.0090.012
2020-01-296.0449.6390.0230.028
2020-02-058.73712.1160.0310.028
2020-02-129.01612.3600.0310.028
2020-02-199.53012.7710.0430.053
2020-02-2612.17715.7650.0590.094
2020-03-0418.77422.0530.1060.101
2020-03-1133.52031.8760.7260.323
2020-03-1863.10452.3901.5292.205
2020-03-2573.84059.8531.7732.728
2020-04-0174.82461.7251.8352.757
2020-04-0875.07263.1591.8442.778
2020-04-1574.95863.2411.8392.779
2020-04-2274.76363.3051.8332.769
2020-04-2974.41963.6981.8242.746
2020-05-0673.23363.0341.7202.704
2020-05-1370.65261.6121.4652.654
2020-05-2070.16861.7811.4422.630
2020-05-2769.11962.2411.3922.440
2020-06-0367.79762.1621.3722.332
2020-06-1067.56462.0961.3612.193
2020-06-1765.78961.1701.3451.877
2020-06-2464.96460.9461.3151.830
2020-07-0164.35361.4111.3601.855
2020-07-0863.07560.4851.3871.857
2020-07-1562.72560.2821.2611.870
2020-07-2262.21459.9920.9731.875
2020-07-2962.98660.8580.9781.897
2020-08-0563.72261.3800.9811.923
2020-08-1263.12561.0200.9861.951
2020-08-1963.17061.0930.9751.970
2020-08-2663.04560.6600.9621.971
2020-09-0261.79759.3670.9181.958
2020-09-0960.59558.5400.9161.944
2020-09-1660.07958.3320.9411.951
2020-09-2359.92158.2230.9341.937
2020-09-3059.96958.1710.9221.924
2020-10-0759.72357.8940.9881.905
2020-10-1460.58558.5261.4381.898
2020-10-2163.33460.4901.5821.918
2020-10-2863.13960.4151.3241.913
2020-11-0463.82060.8461.1502.081
2020-11-1165.01761.4021.2192.090
2020-11-1868.76064.0971.5342.425
2020-11-2569.20764.4311.5332.421
2020-12-0266.70663.0891.3721.989
2020-12-0967.41264.1951.5231.995
2020-12-1669.37666.8471.7092.204
2020-12-2354.81752.8301.4041.805
\n", "
" ], "text/plain": [ " si ci hi wi\n", "date \n", "2020-01-01 0.178 0.359 0.000 0.000\n", "2020-01-08 0.273 0.448 0.000 0.000\n", "2020-01-15 0.722 1.348 0.000 0.000\n", "2020-01-22 2.906 5.626 0.009 0.012\n", "2020-01-29 6.044 9.639 0.023 0.028\n", "2020-02-05 8.737 12.116 0.031 0.028\n", "2020-02-12 9.016 12.360 0.031 0.028\n", "2020-02-19 9.530 12.771 0.043 0.053\n", "2020-02-26 12.177 15.765 0.059 0.094\n", "2020-03-04 18.774 22.053 0.106 0.101\n", "2020-03-11 33.520 31.876 0.726 0.323\n", "2020-03-18 63.104 52.390 1.529 2.205\n", "2020-03-25 73.840 59.853 1.773 2.728\n", "2020-04-01 74.824 61.725 1.835 2.757\n", "2020-04-08 75.072 63.159 1.844 2.778\n", "2020-04-15 74.958 63.241 1.839 2.779\n", "2020-04-22 74.763 63.305 1.833 2.769\n", "2020-04-29 74.419 63.698 1.824 2.746\n", "2020-05-06 73.233 63.034 1.720 2.704\n", "2020-05-13 70.652 61.612 1.465 2.654\n", "2020-05-20 70.168 61.781 1.442 2.630\n", "2020-05-27 69.119 62.241 1.392 2.440\n", "2020-06-03 67.797 62.162 1.372 2.332\n", "2020-06-10 67.564 62.096 1.361 2.193\n", "2020-06-17 65.789 61.170 1.345 1.877\n", "2020-06-24 64.964 60.946 1.315 1.830\n", "2020-07-01 64.353 61.411 1.360 1.855\n", "2020-07-08 63.075 60.485 1.387 1.857\n", "2020-07-15 62.725 60.282 1.261 1.870\n", "2020-07-22 62.214 59.992 0.973 1.875\n", "2020-07-29 62.986 60.858 0.978 1.897\n", "2020-08-05 63.722 61.380 0.981 1.923\n", "2020-08-12 63.125 61.020 0.986 1.951\n", "2020-08-19 63.170 61.093 0.975 1.970\n", "2020-08-26 63.045 60.660 0.962 1.971\n", "2020-09-02 61.797 59.367 0.918 1.958\n", "2020-09-09 60.595 58.540 0.916 1.944\n", "2020-09-16 60.079 58.332 0.941 1.951\n", "2020-09-23 59.921 58.223 0.934 1.937\n", "2020-09-30 59.969 58.171 0.922 1.924\n", "2020-10-07 59.723 57.894 0.988 1.905\n", "2020-10-14 60.585 58.526 1.438 1.898\n", "2020-10-21 63.334 60.490 1.582 1.918\n", "2020-10-28 63.139 60.415 1.324 1.913\n", "2020-11-04 63.820 60.846 1.150 2.081\n", "2020-11-11 65.017 61.402 1.219 2.090\n", "2020-11-18 68.760 64.097 1.534 2.425\n", "2020-11-25 69.207 64.431 1.533 2.421\n", "2020-12-02 66.706 63.089 1.372 1.989\n", "2020-12-09 67.412 64.195 1.523 1.995\n", "2020-12-16 69.376 66.847 1.709 2.204\n", "2020-12-23 54.817 52.830 1.404 1.805" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "def correct_date(y, freq):\n", " \"\"\"\n", " Correct dates and adjust data of last days to avoid last week with < 7 days\n", " or of the last day of February if leap year.\n", " \"\"\"\n", " # numerical columns\n", " cols_num = y.select_dtypes(include='number').columns.to_list()\n", " ndays_lastweek = 7 + y.index[-1].dayofyear % 7\n", " if freq == '7d' and ndays_lastweek > 7:\n", " ts = pd.Timestamp(y.index[-1].date() - pd.to_timedelta(ndays_lastweek-7, unit='D'))\n", " y.index = y.index.where(y.index <= ts, ts) \n", " grouper = pd.Grouper(axis=0, freq=freq)\n", " y = y.groupby(grouper).mean()\n", " # correct the divisor if the last week doesn't have 7 days\n", " y.loc[y.index[-1], cols_num] = y.loc[y.index[-1], cols_num] * (7 / ndays_lastweek)\n", " elif freq == 'd' and y.index.unique()[59].day == 29: # leap year\n", " grouper = pd.Grouper(axis=0, freq=freq)\n", " y = y.groupby(grouper).mean()\n", " # February 28 as the average February 28 and 29\n", " y.loc[y.index[58], cols_num] = y.loc[y.index[58:60], cols_num].mean(axis=0) \n", " y = y.drop(index=y.index[59])\n", " elif freq == 'm' and y.index.unique()[59].day == 29: # just resample data\n", " grouper = pd.Grouper(axis=0, freq=freq)\n", " y = y.groupby(grouper).mean()\n", " return y\n", "\n", "c19idx = correct_date(c19idx, '7d' if freq == 'w' else freq)\n", "c19idx" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Running activity" ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:04.619779Z", "start_time": "2022-03-12T03:51:04.519814Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Year: 2019\n", " Number of days: 365\n", " Number of athletes: 36412\n", " Number of activities: 13290380\n", " Number of actual running activities: 4677389\n", "Year: 2020\n", " Number of days: 365\n", " Number of athletes: 35083\n", " Number of activities: 13290380\n", " Number of actual running activities: 4575794\n", "Total number of actual running activities in 2019 and 2020: 9253183\n" ] } ], "source": [ "for year in years:\n", " print('Year: {}'.format(year))\n", " print(' Number of days:', dfs[year, 'd'].index.value_counts().size)\n", " print(' Number of athletes:', nruns[year, 'd'].size)\n", " print(' Number of activities:', dfs[year, 'd'].shape[0])\n", " print(' Number of actual running activities:', nathletes[year, 'd'].sum())\n", "\n", "nathlete = nruns['2019', 'd'].size\n", "nactivity = nathletes['2019', 'd'].sum() + nathletes['2020', 'd'].sum()\n", "\n", "print('Total number of actual running activities in 2019 and 2020:', nactivity)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Number of running activities per period" ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:04.624516Z", "start_time": "2022-03-12T03:51:04.620753Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Median and interquartile range of number of days the athletes ran each year:\n", "In 2019: 120 (63, 183) days\n", "In 2020: 119 (59, 191) days\n" ] } ], "source": [ "print('Median and interquartile range of number of days the athletes ran each year:')\n", "for year in years:\n", " print('In {}: {:.0f} ({:.0f}, {:.0f}) days'.\n", " format(year, *np.percentile(nruns[year, 'd'], [50, 25, 75]).round()))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Plot of number of athletes per period" ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:04.901880Z", "start_time": "2022-03-12T03:51:04.625423Z" } }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "fig, axs = plt.subplots(3, 1, squeeze=True, sharex=True, figsize=(12, 7))\n", " \n", "for i, year in enumerate(years):\n", " y = nathletes[year, 'd'].values\n", " axs[0].plot(np.linspace(0, 365, 366), np.append(y, y[-1]),\n", " color=colors[i], lw=2, label=year, drawstyle='steps-post')\n", " y = nathletes[year, 'w'].values\n", " axs[1].plot(np.linspace(0, 365, 53), np.append(y, y[-1]),\n", " color=colors[i], lw=3, label=year, drawstyle='steps-post')\n", " x = nathletes[year, 'm'].index.day_of_year.values\n", " y = nathletes[year, 'm'].values\n", " axs[2].plot(np.append([0], x), np.append(y, y[-1]),\n", " color=colors[i], lw=3, label=year, drawstyle='steps-post')\n", "\n", "axs[0].set_xlim([0, 365])\n", "axs[0].legend(['2019', '2020'])\n", "axs[0].set_ylabel('Per day')\n", "axs[1].set_ylabel('Per week')\n", "axs[2].set_ylabel('Per month')\n", "axs[2].set_xlabel('Month')\n", "axs[2].set_xticks(np.append([0], x[:-1]))\n", "axs[2].set_xticklabels(months, rotation=0, ha='center')\n", "offset = matplotlib.transforms.ScaledTranslation(34/72, 0, fig.dpi_scale_trans)\n", "for label in axs[2].get_xticklabels():\n", " label.set_transform(label.get_transform() + offset)\n", "for ax in axs:\n", " ax.tick_params(axis='both', which='major', labelsize=13)\n", "plt.suptitle('Number of athletes running per period (out of a total of {:,d})'\n", " .format(nruns['2019', 'd'].size), fontsize=18)\n", "plt.tight_layout(h_pad=.5)\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 13, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:04.908483Z", "start_time": "2022-03-12T03:51:04.902979Z" } }, "outputs": [ { "data": { "text/plain": [ "count 36412.000\n", "mean 128.457\n", "std 81.522\n", "min 1.000\n", "25% 63.000\n", "50% 120.000\n", "75% 183.000\n", "max 365.000\n", "dtype: float64" ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "nruns['2019', 'd'].describe()" ] }, { "cell_type": "code", "execution_count": 14, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:04.914757Z", "start_time": "2022-03-12T03:51:04.909461Z" } }, "outputs": [ { "data": { "text/plain": [ "count 0.010\n", "mean 2.841\n", "std 4.477\n", "min 365.000\n", "25% 5.794\n", "50% 3.042\n", "75% 1.995\n", "max 1.000\n", "dtype: float64" ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" } ], "source": [ "365 / nruns['2019', 'd'].describe()" ] }, { "cell_type": "code", "execution_count": 15, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:04.920314Z", "start_time": "2022-03-12T03:51:04.915803Z" } }, "outputs": [ { "data": { "text/plain": [ "count 35083.000\n", "mean 130.428\n", "std 87.492\n", "min 1.000\n", "25% 59.000\n", "50% 119.000\n", "75% 191.000\n", "max 365.000\n", "dtype: float64" ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "nruns['2020', 'd'].describe()" ] }, { "cell_type": "code", "execution_count": 16, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:04.925983Z", "start_time": "2022-03-12T03:51:04.921299Z" } }, "outputs": [ { "data": { "text/plain": [ "count 0.010\n", "mean 2.798\n", "std 4.172\n", "min 365.000\n", "25% 6.186\n", "50% 3.067\n", "75% 1.911\n", "max 1.000\n", "dtype: float64" ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "365 / nruns['2020', 'd'].describe()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Volume of training - annual data" ] }, { "cell_type": "code", "execution_count": 17, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:04.929769Z", "start_time": "2022-03-12T03:51:04.926798Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Parameters:\n", " years = ['2019', '2020']\n", " frequency = w\n", " variables = ['distance', 'duration']\n", " estimate = \n" ] } ], "source": [ "print('Parameters:')\n", "print(' years =', years)\n", "print(' frequency =', freq)\n", "print(' variables =', variables)\n", "print(' estimate =', estimate)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Data description of annual data of all subjects" ] }, { "cell_type": "code", "execution_count": 18, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:06.576131Z", "start_time": "2022-03-12T03:51:04.930676Z" } }, "outputs": [], "source": [ "# Statistics for annual data\n", "dfs_stat = dict()\n", "for y in years:\n", " for f in freqs:\n", " idx = pd.Series(data=True, index=dfs[y, f].index)\n", " dfs_stat[y, f] = dfs[y, f][['distance', 'duration']].describe().T" ] }, { "cell_type": "code", "execution_count": 19, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:06.926662Z", "start_time": "2022-03-12T03:51:06.577121Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "2019\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
25%50%75%countmaxmeanminstd
datetime
2019-12-31distance1.6321.60046.181.893e+06711.1429.2410.030.105
duration10.75122.917254.551.893e+068239.20160.5820.0166.521
\n", "
" ], "text/plain": [ " 25% 50% 75% count max mean min \\\n", "datetime \n", "2019-12-31 distance 1.63 21.600 46.18 1.893e+06 711.14 29.241 0.0 \n", " duration 10.75 122.917 254.55 1.893e+06 8239.20 160.582 0.0 \n", "\n", " std \n", "datetime \n", "2019-12-31 distance 30.105 \n", " duration 166.521 " ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
distanceduration
datetime
2019-12-315.537e+073.040e+08
\n", "
" ], "text/plain": [ " distance duration\n", "datetime \n", "2019-12-31 5.537e+07 3.040e+08" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "2020\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
25%50%75%countmaxmeanminstd
datetime
2020-12-31distance0.018.06043.0201.893e+06624.5727.0740.030.320
duration0.0103.683239.4331.893e+066913.00149.8580.0167.785
\n", "
" ], "text/plain": [ " 25% 50% 75% count max mean min \\\n", "datetime \n", "2020-12-31 distance 0.0 18.060 43.020 1.893e+06 624.57 27.074 0.0 \n", " duration 0.0 103.683 239.433 1.893e+06 6913.00 149.858 0.0 \n", "\n", " std \n", "datetime \n", "2020-12-31 distance 30.320 \n", " duration 167.785 " ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
distanceduration
datetime
2020-12-315.126e+072.837e+08
\n", "
" ], "text/plain": [ " distance duration\n", "datetime \n", "2020-12-31 5.126e+07 2.837e+08" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "grouper = pd.Grouper(axis=0, freq='y', sort=True)\n", "for year in years:\n", " print(year)\n", " display(dfs[year, freq][['distance', 'duration']\n", " ].groupby(grouper).describe().stack(level=0))\n", " display(dfs[year, freq][['distance', 'duration']\n", " ].groupby(grouper).sum()) " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Statistics for annual data considering only the actual runs" ] }, { "cell_type": "code", "execution_count": 20, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:07.186780Z", "start_time": "2022-03-12T03:51:06.927638Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "2019\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
25%50%75%countmaxmeanminstd
datetime
2019-12-31distance16.3832.6854.2401.425e+06711.1438.8530.00928.823
duration93.85182.95296.6331.425e+068239.20213.3650.017159.943
\n", "
" ], "text/plain": [ " 25% 50% 75% count max mean \\\n", "datetime \n", "2019-12-31 distance 16.38 32.68 54.240 1.425e+06 711.14 38.853 \n", " duration 93.85 182.95 296.633 1.425e+06 8239.20 213.365 \n", "\n", " min std \n", "datetime \n", "2019-12-31 distance 0.009 28.823 \n", " duration 0.017 159.943 " ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
distanceduration
datetime
2019-12-315.537e+073.040e+08
\n", "
" ], "text/plain": [ " distance duration\n", "datetime \n", "2019-12-31 5.537e+07 3.040e+08" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "2020\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
25%50%75%countmaxmeanminstd
datetime
2020-12-31distance15.8431.56053.8791.332e+06624.5738.4910.01029.453
duration91.00177.852296.0001.332e+066913.00213.0470.033162.972
\n", "
" ], "text/plain": [ " 25% 50% 75% count max mean \\\n", "datetime \n", "2020-12-31 distance 15.84 31.560 53.879 1.332e+06 624.57 38.491 \n", " duration 91.00 177.852 296.000 1.332e+06 6913.00 213.047 \n", "\n", " min std \n", "datetime \n", "2020-12-31 distance 0.010 29.453 \n", " duration 0.033 162.972 " ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
distanceduration
datetime
2020-12-315.126e+072.837e+08
\n", "
" ], "text/plain": [ " distance duration\n", "datetime \n", "2020-12-31 5.126e+07 2.837e+08" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "grouper = pd.Grouper(axis=0, freq='y', sort=True)\n", "for year in years:\n", " print(year)\n", " display(dfs_no0[year, freq][['distance', 'duration']\n", " ].groupby(grouper).describe().stack(level=0)) \n", " display(dfs_no0[year, freq][['distance', 'duration']\n", " ].groupby(grouper).sum())" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Plot of annual number of athletes and running volume per period" ] }, { "cell_type": "code", "execution_count": 21, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:07.564706Z", "start_time": "2022-03-12T03:51:07.187712Z" }, "scrolled": false }, "outputs": [ { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
year20192020
varN athletesDuration (min)Distance (km)Pace (min/km)N athletesDuration (min)Distance (km)Pace (min/km)
127314.000000210.27611538.5931415.44853627436.000000206.78770237.8281805.466499
827312.000000216.29957939.9056835.42027027241.000000210.64583438.7029745.442627
1527061.000000218.45058640.2766255.42375626881.000000214.58602239.4635625.437574
2226878.000000220.01995140.7160685.40376226737.000000215.76885039.9037185.407237
2926919.000000217.57014740.2382505.40704827040.000000217.18859540.1114665.414626
3627244.000000224.06821141.6103955.38491026567.000000218.13568040.3100685.411444
4327329.000000225.38598641.8792715.38180326500.000000220.85190340.8403005.407695
5027400.000000225.70124241.9205355.38402626706.000000219.89832240.6743005.406321
5727557.000000231.78164743.0281535.38674426717.000000218.65806040.4930025.399898
6427306.000000225.82135042.0982755.36414726829.000000219.45991040.5666015.409867
7127838.000000228.73712642.5776675.37223326101.000000201.48411437.1170605.428343
7828065.000000234.72239143.7286305.36770525944.000000204.89669937.6051075.448640
8528114.000000225.63934042.1002805.35956925997.000000208.15905038.4618065.412098
9228469.000000217.95009740.2907785.40942926365.000000213.87077839.4059055.427379
9928172.000000225.72454341.7411995.40771626407.000000213.75303039.3009595.438876
10626890.000000183.45697233.0463815.55150026806.000000218.72838240.1960115.441545
11328012.000000207.29364637.5857035.51522626838.000000216.70055839.7174235.456058
12027151.000000193.50977434.8697155.54950827621.000000219.92735740.1327795.479993
12727391.000000191.07137134.5228225.53463927460.000000219.35843040.0272855.480223
13427856.000000197.35122435.6790525.53129127530.000000223.42551440.5100395.515312
14127493.000000198.81201535.6884415.57076827411.000000224.50200640.3908545.558239
14827415.000000195.99181235.2547095.55930927401.000000222.89461139.9894515.573835
15527789.000000202.64144336.1665185.60301227352.000000218.70074939.4558185.542928
16227679.000000201.53942336.0916525.58410127055.000000222.92259540.0352125.568163
16927792.000000206.47758936.7466895.61894426760.000000221.19599339.4015015.613898
17627504.000000207.88240236.7259585.66036726350.000000219.50785438.9891545.629972
18327904.000000211.46875437.5770935.62759725496.000000217.39641238.5470425.639769
19028089.000000218.81807538.8504605.63231625415.000000218.54567638.6631875.652552
19727907.000000218.22111838.7038725.63822525171.000000218.04885938.4081355.677153
20427990.000000226.45851740.1949475.63400524835.000000216.70289237.9461455.710801
21128281.000000229.50549540.8479345.61853424827.000000216.62099737.9543585.707408
21828435.000000232.10729341.6141895.57760024641.000000215.52928737.5728605.736302
22528651.000000237.79626042.4835245.59737624655.000000216.26430937.9199235.703184
23228784.000000237.59237842.7628705.55604424498.000000214.26732237.7575025.674828
23928787.000000241.00997343.3293735.56227724722.000000212.52466637.7189435.634428
24628994.000000240.46380243.6508025.50880624687.000000212.47820337.6429705.644565
25328857.000000238.19890843.2972615.50147824492.000000209.06904437.1256165.631396
26028687.000000228.73159341.6242925.49514724570.000000208.13085237.0178605.622444
26728857.000000241.72696944.0216115.49109824467.000000205.25568836.7250605.588982
27426878.000000207.72157837.9438445.47444724481.000000213.59027337.8955435.636290
28127926.000000220.64562840.2935265.47595724275.000000205.65597036.6534985.610814
28826817.000000198.83700136.3922195.46372324333.000000204.78393736.7411245.573698
29526954.000000196.60049935.8243365.48790324223.000000204.12197436.5817725.579882
30227159.000000211.77990538.3512695.52211024016.000000202.31838536.3748295.562044
30925471.000000185.54920134.0865875.44346724460.000000202.99122136.5330165.556377
31625655.000000186.57675234.1648015.46108124131.000000203.07511036.7300225.528859
32325867.000000185.83267434.3208945.41456424132.000000204.21598037.0050855.518592
33025845.000000184.16085833.7410955.45805824199.000000204.32973036.9519065.529613
33725253.000000186.09357434.2486205.43360823561.000000202.80475036.8297275.506550
34424416.000000183.53550133.8136435.42785423256.000000202.65152636.7824415.509464
35124274.000000187.00121434.5345285.41490622619.000000202.15141536.6258715.519361
35826334.000000195.80697735.8133905.46742423627.000000198.71653135.6863905.568412
36626334.000000195.80697735.8133905.46742423627.000000198.71653135.6863905.568412
\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "for year in years:\n", " y = dfs[year, freq][['duration', 'distance']].groupby(dfs[year, freq].index).agg('sum')\n", " y['pace'] = y['duration'] / y['distance']\n", " y['nathletes'] = nathletes[year, freq]\n", " y.loc[:, ['duration', 'distance']] = y.loc[:, ['duration', 'distance']\n", " ].div(y['nathletes'], axis=0)\n", " offset = 30 if freq == 'm' else 0\n", " y.index = dfs['2019', freq].index.dayofyear.unique() - offset \n", " y.index.rename('Period', inplace=True)\n", " last = y.iloc[-1]\n", " last.name = 366\n", " y = pd.concat([y, last.to_frame().transpose()]) # for plotting with steps\n", " y = y[['nathletes', 'duration', 'distance', 'pace']]\n", " y.rename(columns={'nathletes': 'N athletes', 'distance': 'Distance (km)',\n", " 'duration': 'Duration (min)', 'pace': 'Pace (min/km)'}, inplace=True)\n", " if year == '2019':\n", " data = y.copy(deep=True)\n", " else:\n", " data = pd.concat([data, y], axis=1, keys=years, names=['year', 'var'])\n", " \n", "display_df(data, subset=data.columns)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Plot of number of athletes and running volume from only actual runs" ] }, { "cell_type": "code", "execution_count": 22, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:08.045780Z", "start_time": "2022-03-12T03:51:07.565728Z" } }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "fig, axs = plt.subplots(4, 1, sharex=True, figsize=(12, 8))\n", "cols = data['2019'].columns\n", "precision = [0, 1, 1, 2]\n", "for a, ax in enumerate(axs):\n", " for y, year in enumerate(years):\n", " ax.plot(np.linspace(0, 365, data[year].shape[0]), data[year][cols[a]],\n", " color=colors[y], lw=3, label=year, drawstyle='steps-post')\n", " msd = '{:g}±{:g}'.format(np.round(data[year][cols[a]].iloc[:-1].mean(), precision[a]),\n", " np.round(data[year][cols[a]].iloc[:-1].std(), precision[a]))\n", " ax.text(.98, .85-y/7, msd, ha='right', c=colors[y],\n", " fontsize=14, transform=ax.transAxes)\n", " ax.margins(x=0, y=0.3)\n", " ax.tick_params(axis='both', which='major', labelsize=13)\n", " ax.set_ylabel(cols[a])\n", "\n", "axs[0].text(.97, 1.03, 'Mean±SD', ha='right', c='k',\n", " fontsize=14, transform=axs[0].transAxes)\n", "axs[0].set_xlim([0, 365])\n", "axs[0].legend(['2019', '2020'], loc='lower left', frameon=True, framealpha=.5, fontsize=14)\n", "axs[3].set_xlabel('Month')\n", "x = nathletes['2019', 'm'].index.day_of_year.values\n", "axs[3].set_xticks(np.append([0], x[:-1]))\n", "axs[3].set_xticklabels(months, rotation=0, ha='center')\n", "offset = matplotlib.transforms.ScaledTranslation(34/72, 0, fig.dpi_scale_trans)\n", "for label in axs[3].get_xticklabels():\n", " label.set_transform(label.get_transform() + offset)\n", "\n", "plt.suptitle('Number of athletes and running volume per athlete at each {} (from only actual runs)'\n", " .format(freq_s), fontsize=18)\n", "fig.align_ylabels(axs)\n", "plt.tight_layout(h_pad=.05)\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Data for relative difference between 2020 and 2019 (in %)\n", "\n", "The difference between 2020 and 2019 in the average running volume per period (day, week or month) will be normalized by the 2019 average running volume per period (a single number) because the year average running volume per period will also be shown in the plots and it will be easier to understand the data.\n", "\n", "The difference between 2020 and 2019 in the number of athletes per period (day, week or month) can be normalized by the total number of athletes in 2019 or by the average number of athletes per period in 2019. We will adopt the second option because it seems more reasonable to compare the differences in weekly running volume with the differences in the number of athletes who actually ran on that week. Also, when calculating this relative difference, the numbers in the numerator and denominator will be equivalent. \n", "However, because we show in the plots the total number of athletes in each year (after all, the mean values are based on the data of all athletes), it will not be possible to directly relate the plotted relative difference in the number of athletes with the total number of athletes in that year.\n", "\n", "Be aware." ] }, { "cell_type": "code", "execution_count": 23, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:08.646579Z", "start_time": "2022-03-12T03:51:08.046745Z" }, "scrolled": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
athletedistancedurationcountrydistance_dduration_d
datetime
2020-01-0100.0000.000United States0.0000.000
2020-01-01170.330394.200Germany222.493226.676
2020-01-01214.65079.067United Kingdom18.296-11.790
2020-01-01341.410195.667United Kingdom-211.070-160.500
2020-01-01441.340209.100United States22.81014.603
.....................
2020-12-2337594128.154572.587United Kingdom394.369317.038
2020-12-233759520.05187.461United States-450.299-334.071
2020-12-2337596144.636625.774United States372.299294.362
2020-12-23375970.0000.000United States-109.220-109.896
2020-12-2337598102.052847.220China191.423326.483
\n", "

1893424 rows × 6 columns

\n", "
" ], "text/plain": [ " athlete distance duration country distance_d duration_d\n", "datetime \n", "2020-01-01 0 0.000 0.000 United States 0.000 0.000\n", "2020-01-01 1 70.330 394.200 Germany 222.493 226.676\n", "2020-01-01 2 14.650 79.067 United Kingdom 18.296 -11.790\n", "2020-01-01 3 41.410 195.667 United Kingdom -211.070 -160.500\n", "2020-01-01 4 41.340 209.100 United States 22.810 14.603\n", "... ... ... ... ... ... ...\n", "2020-12-23 37594 128.154 572.587 United Kingdom 394.369 317.038\n", "2020-12-23 37595 20.051 87.461 United States -450.299 -334.071\n", "2020-12-23 37596 144.636 625.774 United States 372.299 294.362\n", "2020-12-23 37597 0.000 0.000 United States -109.220 -109.896\n", "2020-12-23 37598 102.052 847.220 China 191.423 326.483\n", "\n", "[1893424 rows x 6 columns]" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "nathletesd = dict()\n", "for frq in freqs:\n", " # 2019 total number of athletes\n", " Na = nruns['2019', 'd'].size\n", " # 2019 average number of athletes per period (day, week or month)\n", " #Na = estimate(nathletes['2019', frq])\n", " # 2019 average running volume per period (day, week or month) \n", " Vr = estimate(dfs['2019', frq][variables].values, axis=0)\n", " \n", " dfs['2020', frq][vars_d] = (100 * (dfs['2020', frq][variables].values -\n", " dfs['2019', frq][variables].values) / Vr)\n", "\n", " nathletesd[frq] = pd.Series(100 * ((nathletes['2020', frq].values -\n", " nathletes['2019', frq].values\n", " ) / estimate(nathletes['2019', frq])),\n", " index=nathletes['2020', frq].index)\n", " \n", "display(dfs['2020', freq])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Data description" ] }, { "cell_type": "code", "execution_count": 24, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:08.650682Z", "start_time": "2022-03-12T03:51:08.647425Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Relative difference in total number of athletes between years\n" ] }, { "data": { "text/plain": [ "-3.6498956388003956" ] }, "execution_count": 24, "metadata": {}, "output_type": "execute_result" } ], "source": [ "print('Relative difference in total number of athletes between years')\n", "100 * (nruns['2020', 'd'].size - nruns['2019', 'd'].size) / nruns['2019', 'd'].size" ] }, { "cell_type": "code", "execution_count": 25, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:08.666083Z", "start_time": "2022-03-12T03:51:08.651522Z" }, "scrolled": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Relative difference in number of athletes per week between years\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
countmeanstdmin25%50%75%max
052.0-6.5395.291-16.019-9.97-6.107-2.1431.715
\n", "
" ], "text/plain": [ " count mean std min 25% 50% 75% max\n", "0 52.0 -6.539 5.291 -16.019 -9.97 -6.107 -2.143 1.715" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
 mean
datetime 
2020-01-010.445186
2020-01-08-0.259084
2020-01-15-0.656832
2020-01-22-0.514518
2020-01-290.441537
2020-02-05-2.470418
2020-02-12-3.025076
2020-02-19-2.532452
2020-02-26-3.065216
2020-03-04-1.740605
2020-03-11-6.338428
2020-03-18-7.739670
2020-03-25-7.725074
2020-04-01-7.677636
2020-04-08-6.440602
2020-04-15-0.306522
2020-04-22-4.284004
2020-04-291.715061
2020-05-060.251786
2020-05-13-1.189596
2020-05-20-0.299223
2020-05-27-0.051087
2020-06-03-1.594642
2020-06-10-2.277017
2020-06-17-3.765837
2020-06-24-4.211023
2020-07-01-8.786952
2020-07-08-9.757604
2020-07-15-9.983846
2020-07-22-11.512805
2020-07-29-12.603876
2020-08-05-13.844558
2020-08-12-14.581670
2020-08-19-15.639899
2020-08-26-14.833455
2020-09-02-15.716529
2020-09-09-15.928175
2020-09-16-15.023207
2020-09-23-16.019402
2020-09-30-8.746812
2020-10-07-13.322742
2020-10-14-9.064281
2020-10-21-9.965601
2020-10-28-11.469016
2020-11-04-3.689206
2020-11-11-5.561177
2020-11-18-6.331130
2020-11-25-6.006363
2020-12-02-6.174220
2020-12-09-4.232917
2020-12-16-6.039205
2020-12-23-9.878023
\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "print('Relative difference in number of athletes per {} between years'.format(freq_s))\n", "display(nathletesd[freq].describe().to_frame().T) \n", "display_df(nathletesd[freq].groupby('datetime').agg(['mean']))\n" ] }, { "cell_type": "code", "execution_count": 26, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:08.824023Z", "start_time": "2022-03-12T03:51:08.667016Z" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
25%50%75%countmaxmeanminstd
datetime
2020-12-31distance_d-55.1960.040.3881.893e+061848.542-7.411-2385.349100.692
duration_d-56.0360.042.8031.893e+063933.133-6.678-4732.122107.428
\n", "
" ], "text/plain": [ " 25% 50% 75% count max mean \\\n", "datetime \n", "2020-12-31 distance_d -55.196 0.0 40.388 1.893e+06 1848.542 -7.411 \n", " duration_d -56.036 0.0 42.803 1.893e+06 3933.133 -6.678 \n", "\n", " min std \n", "datetime \n", "2020-12-31 distance_d -2385.349 100.692 \n", " duration_d -4732.122 107.428 " ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "grouper = pd.Grouper(axis=0, freq='y', sort=True)\n", "display(dfs['2020', freq][vars_d].groupby(grouper).describe().stack(level=0)) " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Effect size, confidence interval and hypothesis testing" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We will adopt the following rules of thumb for the magnitude of Cohen's d (https://doi.org/10.22237%2Fjmasm%2F1257035100): " ] }, { "cell_type": "code", "execution_count": 27, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:08.830051Z", "start_time": "2022-03-12T03:51:08.824912Z" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
deffect size
00.01Very small
10.20Small
20.50Medium
30.80Large
41.20Very large
52.00Huge
\n", "
" ], "text/plain": [ " d effect size\n", "0 0.01 Very small\n", "1 0.20 Small\n", "2 0.50 Medium\n", "3 0.80 Large\n", "4 1.20 Very large\n", "5 2.00 Huge" ] }, "execution_count": 27, "metadata": {}, "output_type": "execute_result" } ], "source": [ "effect_sizes = pd.DataFrame({'d': [.01, .2, .5, .8, 1.2, 2], 'effect size':\n", " ['Very small', 'Small', 'Medium', 'Large', 'Very large', 'Huge']})\n", "effect_sizes" ] }, { "cell_type": "code", "execution_count": 28, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:08.851469Z", "start_time": "2022-03-12T03:51:08.832897Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Number of comparisons: 52\n", "Alpha level: 0.05\n", "Alpha level with Bonferroni correction: 0.00096\n", "Alpha level with Sidak correction: 0.00099\n" ] } ], "source": [ "n_periods = dfs['2020', freq].index.unique().size\n", "alphaB = alpha / n_periods\n", "alphaS = 1-(1-alpha)**(1/n_periods)\n", "print('Number of comparisons: {}'.format(n_periods))\n", "print('Alpha level: {:.2f}'.format(alpha))\n", "print('Alpha level with Bonferroni correction: {:.5f}'.format(alphaB))\n", "print('Alpha level with Sidak correction: {:.5f}'.format(alphaS))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Statistics for average values per week across athletes" ] }, { "cell_type": "code", "execution_count": 29, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:08.865687Z", "start_time": "2022-03-12T03:51:08.852453Z" } }, "outputs": [], "source": [ "from bootstrap_stat import bootstrap_stat as bp" ] }, { "cell_type": "code", "execution_count": 30, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:23.329882Z", "start_time": "2022-03-12T03:51:08.866565Z" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
distance2019duration2019pace2019nruns2019distance2020duration2020pace2020nruns2020distanceddurationdpacednrunsd
athlete
011.63671.3886.1351.0555.18530.1845.8210.940-22.068-25.668-5.370-4.671
153.140306.4995.7684.35340.601235.2235.7943.203-42.897-44.4000.442-46.708
219.859124.0236.2451.8229.43264.4986.8381.247-35.670-37.08010.142-23.354
329.050138.0894.7542.99271.588312.3174.3634.756145.516108.533-6.68371.619
422.380113.5215.0732.22530.759151.6844.9313.66328.66523.773-2.41458.385
.......................................
37594129.132585.4794.5346.57876.483350.8234.5874.277-180.106-146.1750.906-93.416
37595118.046492.2654.1706.444115.919486.1864.1946.348-7.276-3.7870.412-3.892
3759687.583369.1194.2145.06396.753421.3854.3555.23631.37032.5592.4077.006
3759741.694219.7925.2723.10740.746229.6775.6373.126-3.2446.1586.2470.778
3759879.607588.3237.3903.68294.075659.6807.0124.96749.49644.451-6.46652.157
\n", "

36412 rows × 12 columns

\n", "
" ], "text/plain": [ " distance2019 duration2019 pace2019 nruns2019 distance2020 \\\n", "athlete \n", "0 11.636 71.388 6.135 1.055 5.185 \n", "1 53.140 306.499 5.768 4.353 40.601 \n", "2 19.859 124.023 6.245 1.822 9.432 \n", "3 29.050 138.089 4.754 2.992 71.588 \n", "4 22.380 113.521 5.073 2.225 30.759 \n", "... ... ... ... ... ... \n", "37594 129.132 585.479 4.534 6.578 76.483 \n", "37595 118.046 492.265 4.170 6.444 115.919 \n", "37596 87.583 369.119 4.214 5.063 96.753 \n", "37597 41.694 219.792 5.272 3.107 40.746 \n", "37598 79.607 588.323 7.390 3.682 94.075 \n", "\n", " duration2020 pace2020 nruns2020 distanced durationd paced \\\n", "athlete \n", "0 30.184 5.821 0.940 -22.068 -25.668 -5.370 \n", "1 235.223 5.794 3.203 -42.897 -44.400 0.442 \n", "2 64.498 6.838 1.247 -35.670 -37.080 10.142 \n", "3 312.317 4.363 4.756 145.516 108.533 -6.683 \n", "4 151.684 4.931 3.663 28.665 23.773 -2.414 \n", "... ... ... ... ... ... ... \n", "37594 350.823 4.587 4.277 -180.106 -146.175 0.906 \n", "37595 486.186 4.194 6.348 -7.276 -3.787 0.412 \n", "37596 421.385 4.355 5.236 31.370 32.559 2.407 \n", "37597 229.677 5.637 3.126 -3.244 6.158 6.247 \n", "37598 659.680 7.012 4.967 49.496 44.451 -6.466 \n", "\n", " nrunsd \n", "athlete \n", "0 -4.671 \n", "1 -46.708 \n", "2 -23.354 \n", "3 71.619 \n", "4 58.385 \n", "... ... \n", "37594 -93.416 \n", "37595 -3.892 \n", "37596 7.006 \n", "37597 0.778 \n", "37598 52.157 \n", "\n", "[36412 rows x 12 columns]" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "18aa8167e1cc421ebc35109172acab83", "version_major": 2, "version_minor": 0 }, "text/plain": [ " 0%| | 0/12 [00:00\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
countmeanstdcohensdci_infci_suppvalue
distance201936412.029.23222.977-29.00229.467-
duration201936412.0160.530119.248-159.304161.778-
pace201936412.05.8488.791-5.7905.950-
nruns201936412.02.4641.563-2.4472.480-
distance202036412.027.04424.567-26.79427.303-
duration202036412.0149.699130.758-148.339151.037-
pace202035083.05.94711.550-5.8656.086-
nruns202036412.02.4101.713-2.3922.428-
distanced36412.0-7.48547.9410.156-7.985-6.9850.0
durationd36412.0-6.74748.5470.139-7.238-6.2450.0
paced35083.03.025196.5340.0151.6595.5010.0
nrunsd36412.0-2.17242.3450.051-2.608-1.7550.0
\n", "" ], "text/plain": [ " count mean std cohensd ci_inf ci_sup pvalue\n", "distance2019 36412.0 29.232 22.977 - 29.002 29.467 -\n", "duration2019 36412.0 160.530 119.248 - 159.304 161.778 -\n", "pace2019 36412.0 5.848 8.791 - 5.790 5.950 -\n", "nruns2019 36412.0 2.464 1.563 - 2.447 2.480 -\n", "distance2020 36412.0 27.044 24.567 - 26.794 27.303 -\n", "duration2020 36412.0 149.699 130.758 - 148.339 151.037 -\n", "pace2020 35083.0 5.947 11.550 - 5.865 6.086 -\n", "nruns2020 36412.0 2.410 1.713 - 2.392 2.428 -\n", "distanced 36412.0 -7.485 47.941 0.156 -7.985 -6.985 0.0\n", "durationd 36412.0 -6.747 48.547 0.139 -7.238 -6.245 0.0\n", "paced 35083.0 3.025 196.534 0.015 1.659 5.501 0.0\n", "nrunsd 36412.0 -2.172 42.345 0.051 -2.608 -1.755 0.0" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "data = pd.DataFrame([])\n", "frq = 'd'\n", "for year in years:\n", " y = dfs[year, frq][['athlete', 'distance', 'duration']].groupby('athlete').sum() * 7 /365\n", " y['pace'] = y['duration'] / y['distance']\n", " for col in y.columns:\n", " y.rename(columns={col: col+year}, inplace=True)\n", " data = pd.concat([data, y], axis=1)\n", " data = pd.concat([data, nruns[year, 'd'].to_frame(name='nruns'+year) * 7 / 365], axis=1) \n", "# replace nans with zeros; merging 2019 and 2020 creates nans in nruns2020 due to nonexistent data \n", "data['nruns2020'] = data['nruns2020'].fillna(0)\n", "# replace nans with zeros; merging 2019 and 2020 creates nans in pace2020 due to no runs,\n", "# not replacing nans will use only actual runs to compute pace (makes sense), but it will\n", "# result in 2029 and 2020 having different number of athletes when computing averages.\n", "#data['pace2020'] = data['pace2020'].fillna(0)\n", "# relative differences\n", "# normalize by global values\n", "Vr = estimate(data[['distance2019', 'duration2019', 'pace2019', 'nruns2019']].values, axis=0)\n", "# normalize by individual values\n", "#Vr = data[['distance2019', 'duration2019', 'pace2019', 'nruns2019']].values\n", "\n", "data[['distanced','durationd','paced','nrunsd']] = (100*(data[['distance2020','duration2020',\n", " 'pace2020','nruns2020']].values -\n", " data[['distance2019','duration2019',\n", " 'pace2019','nruns2019']].values) / Vr)\n", "display(data)\n", "\n", "stat = ['count', 'mean', 'std', cohensd]\n", "ys = data.agg(stat, axis=0).T\n", "for var in tqdm(data.columns):\n", " ys.loc[var, ['ci_inf', 'ci_sup', 'pvalue']] = ci_asl(data[var],\n", " alpha2=alpha/2, boot=10000)\n", " if var[-1] != 'd':\n", " ys.loc[var, ['cohensd', 'pvalue']] = ['-', '-']\n", "display(ys)" ] }, { "cell_type": "code", "execution_count": 31, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:37.592506Z", "start_time": "2022-03-12T03:51:23.331289Z" } }, "outputs": [ { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "99d34f44bb7a422c87375c2fd10ee8ab", "version_major": 2, "version_minor": 0 }, "text/plain": [ " 0%| | 0/12 [00:00\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
countmeanstdcohensdci_infci_suppvalue
distance201936412.029.23222.977-28.99629.471-
duration201936412.0160.530119.248-159.312161.729-
pace201936412.05.8488.791-5.7905.952-
nruns201936412.02.4641.563-2.4482.480-
distance202036412.027.04424.567-26.78927.293-
duration202036412.0149.699130.758-148.387151.038-
pace202036412.05.73011.392-5.6505.868-
nruns202036412.02.4101.713-2.3922.428-
distanced36412.0-7.48547.9410.156-7.982-6.9890.0
durationd36412.0-6.74748.5470.139-7.257-6.2740.0
paced36412.0-2.011245.1180.008-4.4180.7050.123
nrunsd36412.0-2.17242.3450.051-2.602-1.7350.0
\n", "" ], "text/plain": [ " count mean std cohensd ci_inf ci_sup pvalue\n", "distance2019 36412.0 29.232 22.977 - 28.996 29.471 -\n", "duration2019 36412.0 160.530 119.248 - 159.312 161.729 -\n", "pace2019 36412.0 5.848 8.791 - 5.790 5.952 -\n", "nruns2019 36412.0 2.464 1.563 - 2.448 2.480 -\n", "distance2020 36412.0 27.044 24.567 - 26.789 27.293 -\n", "duration2020 36412.0 149.699 130.758 - 148.387 151.038 -\n", "pace2020 36412.0 5.730 11.392 - 5.650 5.868 -\n", "nruns2020 36412.0 2.410 1.713 - 2.392 2.428 -\n", "distanced 36412.0 -7.485 47.941 0.156 -7.982 -6.989 0.0\n", "durationd 36412.0 -6.747 48.547 0.139 -7.257 -6.274 0.0\n", "paced 36412.0 -2.011 245.118 0.008 -4.418 0.705 0.123\n", "nrunsd 36412.0 -2.172 42.345 0.051 -2.602 -1.735 0.0" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "data = pd.DataFrame([])\n", "frq = 'd'\n", "for year in years:\n", " y = dfs[year, frq][['athlete', 'distance', 'duration']].groupby('athlete').sum() * 7 /365\n", " y['pace'] = y['duration'] / y['distance']\n", " for col in y.columns:\n", " y.rename(columns={col: col+year}, inplace=True)\n", " data = pd.concat([data, y], axis=1)\n", " data = pd.concat([data, nruns[year, 'd'].to_frame(name='nruns'+year) * 7 / 365], axis=1) \n", "# replace nans with zeros; merging 2019 and 2020 creates nans in nruns2020 due to nonexistent data \n", "data['nruns2020'] = data['nruns2020'].fillna(0)\n", "# replace nans with zeros; merging 2019 and 2020 creates nans in pace2020 due to no runs,\n", "# not replacing nans will use only actual runs to compute pace (makes sense), but it will\n", "# result in 2029 and 2020 having different number of athletes when computing averages.\n", "data['pace2020'] = data['pace2020'].fillna(0)\n", "# relative differences\n", "# normalize by global values\n", "Vr = estimate(data[['distance2019', 'duration2019', 'pace2019', 'nruns2019']].values, axis=0)\n", "# normalize by individual values\n", "#Vr = data[['distance2019', 'duration2019', 'pace2019', 'nruns2019']].values\n", "\n", "data[['distanced','durationd','paced','nrunsd']] = (100*(data[['distance2020','duration2020',\n", " 'pace2020','nruns2020']].values -\n", " data[['distance2019','duration2019',\n", " 'pace2019','nruns2019']].values) / Vr)\n", "\n", "stat = ['count', 'mean', 'std', cohensd]\n", "ys = data.agg(stat, axis=0).T\n", "for var in tqdm(data.columns):\n", " ys.loc[var, ['ci_inf', 'ci_sup', 'pvalue']] = ci_asl(data[var],\n", " alpha2=alpha/2, boot=10000)\n", " if var[-1] != 'd':\n", " ys.loc[var, ['cohensd', 'pvalue']] = ['-', '-']\n", "display(ys)" ] }, { "cell_type": "code", "execution_count": 32, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:37.695962Z", "start_time": "2022-03-12T03:51:37.593824Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Year: 2019\n", " Number of days: 365\n", " Number of athletes with a least one session: 36412\n", " Number of activities: 13290380\n", " Number of actual running activities: 4677389\n", "Year: 2020\n", " Number of days: 365\n", " Number of athletes with a least one session: 35083\n", " Number of activities: 13290380\n", " Number of actual running activities: 4575794\n", "Total number of actual running activities in 2019 and 2020: 9253183\n", "Relative difference in number of athletes [%]: -3.6498956388003956\n" ] } ], "source": [ "for year in years:\n", " print('Year: {}'.format(year))\n", " print(' Number of days:', dfs[year, 'd'].index.value_counts().size)\n", " print(' Number of athletes with a least one session:', nruns[year, 'd'].size)\n", " print(' Number of activities:', dfs[year, 'd'].shape[0])\n", " print(' Number of actual running activities:', nathletes[year, 'd'].sum())\n", "\n", "nathlete = nruns['2019', 'd'].size\n", "nactivity = nathletes['2019', 'd'].sum() + nathletes['2020', 'd'].sum()\n", "\n", "print('Total number of actual running activities in 2019 and 2020:', nactivity)\n", "nd = 100 * (nruns['2020', 'd'].size - nruns['2019', 'd'].size) / nruns['2019', 'd'].size\n", "print('Relative difference in number of athletes [%]: {}'.format(nd))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Number of athletes running per week" ] }, { "cell_type": "code", "execution_count": 33, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:38.271708Z", "start_time": "2022-03-12T03:51:37.697221Z" } }, "outputs": [ { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "411ad2dea22545a88213f0b8ae2f1e66", "version_major": 2, "version_minor": 0 }, "text/plain": [ " 0%| | 0/3 [00:00\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
countmeanstdcohensdci_infci_suppvalue
nathletesw201952.027404.2691075.724-27102.34627679.385-
nathletesw202052.025612.3271365.308-25238.13525984.558-
nathleteswd52.0-6.5395.2911.236-8.056-5.1180.0
\n", "" ], "text/plain": [ " count mean std cohensd ci_inf ci_sup \\\n", "nathletesw2019 52.0 27404.269 1075.724 - 27102.346 27679.385 \n", "nathletesw2020 52.0 25612.327 1365.308 - 25238.135 25984.558 \n", "nathleteswd 52.0 -6.539 5.291 1.236 -8.056 -5.118 \n", "\n", " pvalue \n", "nathletesw2019 - \n", "nathletesw2020 - \n", "nathleteswd 0.0 " ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "nwd = 100 * (nathletes['2020', 'w'].values - nathletes['2019', 'w'].values\n", " ) / nathletes['2019', 'w'].mean()\n", "data = pd.DataFrame(data=np.c_[nathletes['2019', 'w'].values, nathletes['2020', 'w'].values, nwd],\n", " columns=['nathletesw2019', 'nathletesw2020', 'nathleteswd'])\n", "#display(data)\n", "\n", "stat = ['count', 'mean', 'std', cohensd]\n", "ys = data.agg(stat, axis=0).T\n", "for var in tqdm(data.columns):\n", " ys.loc[var, ['ci_inf', 'ci_sup', 'pvalue']] = ci_asl(data[var],\n", " alpha2=alpha/2, boot=1000)\n", " if var[-1] != 'd':\n", " ys.loc[var, ['cohensd', 'pvalue']] = ['-', '-']\n", "display(ys)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Load data with statistics or run bootstrap calculation for time series" ] }, { "cell_type": "code", "execution_count": 34, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T03:51:38.346334Z", "start_time": "2022-03-12T03:51:38.273197Z" }, "scrolled": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "../data/dfsdata_w_all.pkl loaded.\n", "\n" ] }, { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Results for all athletes of country: all countries
 distancedistance2019distance2020durationduration2019duration2020
 ci_infci_supcohensdcountmeannathletespvaluestdci_infci_supcohensdcountmeannathletespvaluestdci_infci_supcohensdcountmeannathletespvaluestdci_infci_supcohensdcountmeannathletespvaluestdci_infci_supcohensdcountmeannathletespvaluestdci_infci_supcohensdcountmeannathletespvaluestd
Period                                                
1-2.499744-0.5671210.01627736412-1.5289320.4451860.001893.92919028.64425029.2553930.9729583641228.950156273140.000029.75478028.19373728.8137080.9491893641228.503074274360.000030.028860-2.186130-0.2042160.01242636412-1.1981030.4451860.018396.417481156.109886159.3766700.99496536412157.735961273140.0000158.534112154.167371157.4578890.97061136412155.812024274360.0000160.529847
8-4.331955-2.3538260.03473736412-3.343197-0.2590840.000096.24199229.62227930.2395760.9922973641229.932550273120.000030.16491728.64381829.2655360.9556303641228.954952272410.000030.299343-3.904785-1.8987610.02957236412-2.896661-0.2590840.000097.952391160.613525163.8901871.01817836412162.242505273120.0000159.345895155.948575159.2416630.97758936412157.590991272410.0000161.203729
15-3.728039-1.7422690.02835036412-2.733606-0.6568320.000096.42470729.61968230.2431410.9854433641229.933147270610.000030.37531028.82103429.4486870.9536913641229.133802268810.000030.548476-3.470117-1.4286930.02469636412-2.449150-0.6568320.000099.171336160.700888164.0171641.00389136412162.350085270610.0000161.720777156.746008160.0926550.97252836412158.417194268810.0000162.892099
22-3.588715-1.5696400.02617836412-2.579107-0.5145180.000098.52333229.73856430.3727720.9694833641230.055105268780.000031.00116628.98259729.6219660.9413093641229.300937267370.000031.127851-3.491327-1.4581500.02508436412-2.474462-0.5145180.000098.646505160.729190164.0767441.00299336412162.410641268780.0000161.926047156.761756160.1131930.97014536412158.437102267370.0000163.312884
29-0.8742991.1452360.001376364120.1353010.4415370.791698.31369429.43169630.0625340.9626423641229.747705269190.000030.90213529.46561930.1081350.9492183641229.787269270400.000031.380864-0.7611201.3076050.002740364120.2737910.4415370.602499.923351159.156875162.5311880.98254736412160.847270269190.0000163.704359159.568903162.9953330.97371836412161.286928270400.0000165.640251
36-6.922447-4.8585450.05873936412-5.890273-2.4704180.0000100.27951430.80757431.4601830.9844313641231.133516272440.000031.62589429.08801429.7371290.9315793641229.411117265670.000031.571258-6.323687-4.2471580.05235236412-5.289857-2.4704180.0000101.044520165.959841169.3369321.01815736412167.651168272440.0000164.661334157.443120160.8693010.95745936412159.156613265670.0000166.228197
43-6.893589-4.8013150.05749236412-5.846568-3.0250760.0000101.69408931.10519531.7585780.9908833641231.432456273290.000031.72167129.39678630.0498370.9322063641229.722837265000.000031.884413-6.287340-4.1975250.05158536412-5.250427-3.0250760.0000101.782639167.455950170.8681211.02111036412169.163287273290.0000165.666113159.014675162.4486570.95849536412160.732051265000.0000167.692196
50-6.904102-4.8002030.05727936412-5.858228-2.5324520.0000102.27539331.21651331.8742880.9867533641231.545168274000.000031.96866329.50556230.1588240.9400623641229.832139267060.000031.734225-6.385927-4.2624720.05158836412-5.329291-2.5324520.0000103.304609168.126954171.5626141.01301236412169.839999274000.0000167.658394159.574571162.9904380.96838136412161.282121267060.0000166.548132
57-10.819771-8.7019210.09434136412-9.755942-3.0652160.0000103.41151132.23579732.8923371.0164143641232.564177275570.000032.03828829.38858230.0349260.9424193641229.711401267170.000031.526745-10.406049-8.2512650.08900936412-9.326309-3.0652160.0000104.779456173.698072177.1546621.04128236412175.414887275570.0000168.460577158.728080162.1563500.96575336412160.438520267170.0000166.127972
64-6.820829-4.6701050.05507136412-5.745462-1.7406050.0000104.32766131.24114031.9043430.9784473641231.570238273060.000032.26565329.56448230.2179180.9382173641229.890183268290.000031.858510-5.837797-3.6897550.04546736412-4.761120-1.7406050.0000104.716100167.635200171.0849621.00524136412169.347407273060.0000168.464440159.980371163.4248170.96278336412161.701909268290.0000167.952683
71-21.387353-19.2749260.19787736412-20.332194-6.3384280.0000102.75152332.22186032.8845481.0039113641232.551826278380.000032.42499926.30274626.9102690.8982683641226.606404261010.000029.619673-20.022895-17.9006330.18296036412-18.960605-6.3384280.0000103.632417173.133746176.6438131.02094436412174.875978278380.0000171.288580142.813773146.0493800.92180936412144.428674261010.0000156.679526
78-24.750265-22.5259250.21941536412-23.631821-7.7396700.0000107.70352233.36116234.0510471.0099033641233.704383280650.000033.37388426.48851227.0984480.8979133641226.794103259440.000029.840412-22.858004-20.6451810.20194636412-21.748251-7.7396700.0000107.693645179.140913182.7047511.03672536412180.915190280650.0000174.506491144.361255147.6287090.91631836412145.991430259440.0000159.323912
85-18.329787-16.1767470.16419836412-17.254536-7.7250740.0000105.08355732.17666532.8378401.0148483641232.505967281140.000032.03038327.14948927.7724870.9055483641227.460496259970.000030.324714-17.026211-14.8612820.15152236412-15.941444-7.7250740.0000105.209044172.494784175.9314481.03972336412174.217962281140.0000167.561879146.985235150.2655650.92702136412148.618884259970.0000160.318761
92-11.223472-9.0752620.09721636412-10.152865-7.6776360.0000104.43563531.19721431.8087731.0625493641231.501652284690.000029.64725128.21658428.8508660.9245883641228.532810263650.000030.860011-10.796075-8.5830180.08978736412-9.682019-7.6776360.0000107.832817168.766663172.0619151.06236536412170.405946284690.0000160.402511153.168115156.5360220.94325236412154.858373263650.0000164.174902
99-14.059822-11.8919310.12355536412-12.971508-6.4406020.0000104.98558831.98581332.6068061.0663053641232.295207281720.000030.28702228.18918628.8174750.9299863641228.502154264070.000030.647922-13.336068-11.1087860.11191736412-12.220322-6.4406020.0000109.190710172.968940176.3377391.05991036412174.643300281720.0000164.771768153.340541156.7031340.94781236412155.019671264070.0000163.555258
10616.64047518.8331460.1653573641217.739292-0.3065220.0000107.27853224.13037824.6796510.9126433641224.404514268900.000026.74046829.27266829.9148480.9470753641229.591735268060.000031.24539714.78696017.0266290.1452213641215.906579-0.3065220.0000109.533847133.955702137.0170560.90848536412135.481653268900.0000149.129257159.309855162.7501780.96249836412161.024745268060.0000167.298753
1130.1535752.2984430.011795364121.228984-4.2840040.0247104.19412128.63121629.2016101.0471293641228.914938280120.000027.61353228.95832229.5930250.9478313641229.274310268380.000030.885566-0.9720521.2822080.001418364120.155632-4.2840040.7866109.779177157.889010161.0837641.02355536412159.472416280120.0000155.802527158.019061161.4207180.96205136412159.722332268380.0000166.022652
12014.16095516.2226510.1505043641215.1924481.7150610.0000100.94410225.71755826.2854880.9408153641226.000978271510.000027.63664130.12381030.7604620.9783343641230.443466276210.000031.11766112.93349915.1302690.1313073641214.0347931.7150610.0000106.885228142.672594145.9228700.91365036412144.292647271510.0000157.929957165.091709168.5847260.98416436412166.829988276210.0000169.514499
12713.40199415.4359420.1461773641214.4199380.2517860.000098.64677325.68940426.2516170.9449343641225.969862273910.000027.48325629.86404930.5081520.9644113641230.186456274600.000031.30042012.41178114.6068040.1262103641213.5100370.2517860.0000107.043951142.092281145.4000170.89193436412143.733822273910.0000161.148449163.679957167.1901870.96754936412165.428499274600.0000170.976943
13410.38089212.4134350.1149843641211.398637-1.1895960.000099.13238727.00997727.5825460.9768823641227.295278278560.000027.94122130.30395930.9532110.9696073641230.628402275300.000031.58848110.02415612.3198450.1001723641211.176270-1.1895960.0000111.571055149.273169152.7076110.90452336412150.978131278560.0000166.914704167.138262170.7262910.96773336412168.925200275300.0000174.557609
14110.78889212.8694390.1175253641211.831203-0.2992230.0000100.66977826.65348227.2411820.9431673641226.946675274930.000028.57040830.07528930.7367710.9478083641230.406286274110.000032.08062210.61569812.9163120.1056853641211.764545-0.2992230.0000111.317372148.411105151.8446450.89955836412150.113664274930.0000166.874994167.157585170.8683180.93808736412169.005397274110.0000180.159670
14811.13052413.1522480.1230213641212.138481-0.0510870.000098.67014326.25745226.8319340.9475593641226.543663274150.000028.01267929.77128230.4186610.9543643641230.093127274010.000031.53213011.46455913.6655560.1170563641212.560390-0.0510870.0000107.302228145.916647149.2178370.91895336412147.564416274150.0000160.578783165.930898169.5674480.94876736412167.734133274010.0000176.791641
1555.9510457.9791760.070325364126.965481-1.5946420.000099.04674827.30479927.8998330.9544863641227.601653277890.000028.91783229.32313629.9580010.9604143641229.638458273520.000030.8600924.8741147.1088750.055144364125.997833-1.5946420.0000108.766352152.929810156.3792180.91964836412154.652396277890.0000168.164781162.522595166.0367670.96131336412164.283832273520.0000170.895200
1626.8956398.9038930.080699364127.905399-2.2770170.000097.96212527.14429827.7259600.9700623641227.435484276790.000028.28219029.42837130.0681650.9528933641229.747135270550.000031.2177206.6366628.8608330.071741364127.743333-2.2770170.0000107.935048151.532853154.8829750.93554636412153.202507276790.0000163.757345163.851347167.4456600.94883036412165.636900270550.0000174.569540
1692.1070754.1166600.031842364123.110641-3.7658370.000097.68959927.74990428.3432750.9708623641228.047456277920.000028.88924328.63760029.2763010.9341303641228.957052267600.000030.9989581.9587074.2302570.027925364123.091777-3.7658370.0000110.716729155.846790159.3554870.92429336412157.597088277920.0000170.505480160.744399164.3690750.92055836412162.561924267600.0000176.590545
1760.5981232.6379270.016356364121.620440-4.2110230.001999.07606227.44206928.0389970.9483233641227.741150275040.000029.25286427.89816828.5351600.9114683641228.214990263500.000030.955555-0.0089052.2827580.010212364121.136192-4.2110230.0516111.262368155.271358158.7993720.90898036412157.025090275040.0000172.748632157.055550160.6694930.90404736412158.849609263500.0000175.709530
183-7.192073-5.1567090.06236236412-6.175773-8.7869520.000099.03097028.49532429.0955770.9858923641228.796858279040.000029.20893726.67567727.3024690.8842203641226.990975254960.000030.525195-7.243808-4.9992320.05573636412-6.124141-8.7869520.0000109.876960160.330956163.7835540.96211036412162.057127279040.0000168.439245150.436466154.0298440.87313436412152.222864254960.0000174.340721
190-11.246202-9.1749100.10108736412-10.203976-9.7576040.0000100.94222929.66077230.2772000.9943403641229.970080280890.000030.14068026.67306627.3040180.8756443641226.986293254150.000030.818804-11.303717-8.9465950.08843636412-10.125389-9.7576040.0000114.493721167.006728170.6144760.95962136412168.800969280890.0000175.903764150.711212154.3814850.85491436412152.541425254150.0000178.428919
197-11.680775-9.6195080.10602236412-10.644657-9.9838460.0000100.40083529.35600629.9727710.9851633641229.663543279070.000030.11029026.23633226.8639600.8697253641226.550895251710.000030.527931-11.449319-9.1144500.09080136412-10.285235-9.9838460.0000113.271832165.445263169.0406970.95376136412167.249718279070.0000175.358163148.926006152.5531260.85517036412150.733490251710.0000176.261492
204-18.217598-16.0958700.16656636412-17.155766-11.5128050.0000102.99695430.58002131.2178260.9992243641230.897961279900.000030.92194325.56903426.1979590.8485653641225.881372248350.000030.500150-17.568872-15.1442990.13847536412-16.362916-11.5128050.0000118.165382172.225813175.9402260.96623436412174.079257279900.0000180.162567145.954974149.6739890.81722536412147.803371248350.0000180.860156
211-21.058883-18.9364890.19350836412-19.998108-12.6038760.0000103.34521831.40447532.0471191.0093673641231.726365282810.000031.43194325.56833226.1946040.8519993641225.878634248270.000030.374032-20.252896-17.8067430.15953136412-19.028129-12.6038760.0000119.275763176.390670180.1517230.97705136412178.255655282810.0000182.442490145.860250149.5686380.81888436412147.699920248270.0000180.367285
218-25.248499-23.1109500.23329436412-24.181205-13.8445580.0000103.65121332.17279232.8233181.0222633641232.497513284350.000031.78977425.11684025.7355130.8415213641225.426586246410.000030.215020-23.273535-20.8304750.18613336412-22.047027-13.8445580.0000118.447877179.399294183.1049091.01201836412181.258126284350.0000179.105673143.998958147.7427560.79972836412145.854585246410.0000182.380241
225-27.589381-25.4366390.25342236412-26.511648-14.5816700.0000104.61455033.09925133.7621181.0357803641233.428415286510.000032.27366725.36620725.9874520.8459653641225.676032246550.000030.351178-26.537989-24.1152910.21487936412-25.330537-14.5816700.0000117.882979185.228267189.0261991.01006436412187.111410286510.0000185.247020144.614042148.2700420.82040636412146.435146246550.0000178.491077
232-29.803073-27.6603360.27524736412-28.730355-15.6398990.0000104.38036233.47287234.1384091.0515093641233.804417287840.000032.14847925.09478225.7120300.8426853641225.403254244980.000030.145622-28.380059-26.0009280.23463136412-27.188447-15.6398990.0000115.877377185.982123189.6700771.04934436412187.818824287840.0000178.986835142.349109145.9905700.81571836412144.159092244980.0000176.726657
239-30.637217-28.5043410.28428336412-29.569226-14.8334550.0000104.01336233.92341034.5913691.0561143641234.255813287870.000032.43572625.30420225.9164970.8607893641225.609351247220.000029.751011-29.965194-27.6214120.25215836412-28.799177-14.8334550.0000114.210915188.665628192.4168961.04025436412190.540319287870.0000183.167032142.555771146.0389600.85323236412144.294045247220.0000169.114639
246-32.670771-30.5128190.30143036412-31.587037-15.7165290.0000104.79055734.42887535.0906641.0812593641234.758084289940.000032.14594625.21575825.8279890.8628713641225.521586246870.000029.577521-30.691649-28.3564500.26046836412-29.528406-15.7165290.0000113.366702189.660624193.2954041.08101036412191.475543289940.0000177.126579142.323361145.7897570.85433736412144.058261246870.0000168.619955
253-33.010826-30.8781710.30648536412-31.946665-15.9281750.0000104.23552833.98485334.6424351.0736483641234.313662288570.000031.95987924.67361525.2708760.8618893641224.972003244920.000028.973579-31.141459-28.8284840.26636936412-29.983823-15.9281750.0000112.564937186.967883190.5879811.06960436412188.775840288570.0000176.491353138.951901142.3182100.85515136412140.627239244920.0000164.447204
260-27.775639-25.6665570.26023536412-26.724611-15.0232070.0000102.69417032.47563533.1145571.0517173641232.793476286870.000031.18090024.68081825.2797150.8575273641224.978821245700.000029.128912-25.920006-23.6125470.21964236412-24.761778-15.0232070.0000112.736764178.421051182.0151911.03440336412180.204966286870.0000174.211529138.726805142.1406700.84506736412140.442026245700.0000166.190323
267-35.972500-33.8647390.34222236412-34.917454-16.0194020.0000102.03151734.57142535.2030311.1327783641234.887719288570.000030.79836124.37942724.9748410.8527103641224.677360244670.000028.939924-34.546947-32.2686480.30191036412-33.410087-16.0194020.0000110.662231189.822535193.3181621.12447836412191.571876288570.0000170.365117136.246249139.6078320.84484236412137.921315244670.0000163.250971
274-9.688850-7.6063540.08521836412-8.653190-8.7468120.0000101.54154427.70510828.3149870.9390423641228.008751268780.000029.82695225.17900225.7822280.8690803641225.478435244810.000029.316558-7.194488-4.9273540.05481736412-6.058329-8.7468120.0000110.518278151.644852155.0203830.93137136412153.332433268780.0000164.630856141.901836145.3227610.85591236412143.603853244810.0000167.778833
281-23.166826-21.0705720.21657436412-22.115603-13.3227420.0000102.11585530.59458031.2098311.0299443641230.902917279260.000030.00446724.13865124.7342480.8394103641224.436001242750.000029.110915-21.161023-18.8342110.17679436412-20.000502-13.3227420.0000113.128977167.507029170.9398681.01457236412169.223053279260.0000166.792580135.389366138.8539390.81216236412137.105863242750.0000168.815827
288-8.692104-6.6884950.07876536412-7.692815-9.0642810.000097.66813426.50710627.0979450.9294583641226.802432268170.000028.83663424.25669324.8488200.8497503641224.552943243330.000028.894303-7.066694-4.8806630.05605136412-5.972256-9.0642810.0000106.550487144.772551148.1168310.89661636412146.441059268170.0000163.326311135.193177138.5346770.84284136412136.850696243330.0000162.368278
295-8.451073-6.4879880.07855336412-7.465578-9.9656010.000095.03813726.22783926.8102210.9348123641226.518982269540.000028.36826424.04159924.6319130.8486413641224.335941242230.000028.676376-7.155538-4.9981410.05794036412-6.066644-9.9656010.0000104.705873143.890838147.1937010.90108236412145.533611269540.0000161.509800134.158886137.4318750.84766136412135.791678242230.0000160.195706
302-16.788075-14.7677190.16146536412-15.778916-11.4690160.000097.72325028.30727528.9017130.9898923641228.605463271590.000028.89755423.70000624.2850040.8421083641223.991484240160.000028.489791-16.355329-14.1876050.14417736412-15.269975-11.4690160.0000105.911147156.307454159.6336810.97692236412157.962497271590.0000161.694107131.823590135.0680090.84204436412133.441677240160.0000158.473579
3091.4189603.3441480.025434364122.383509-3.6892060.000093.71456223.55763524.1299420.8540203641223.844322254710.000027.92009824.24900824.8331740.8616103641224.541294244600.000028.4830523.0822985.0983050.041662364124.088201-3.6892060.000098.126636128.229731131.3511580.86039236412129.795773254710.0000150.856525134.741474137.9930340.86225436412136.360685244600.0000158.144438
316-0.0134411.8558260.010124364120.923626-5.5611770.053891.23362723.78986124.3555250.8727783641224.071679256550.000027.58051524.04835024.6355550.8535373641224.341760241310.000028.5187060.9606392.9360180.020219364121.945900-5.5611770.000196.242442129.917547132.9954220.87410136412131.457393256550.0000150.391536132.984073136.1957580.85872936412134.582157241310.0000156.722467
323-0.4490921.4242050.005395364120.491027-6.3311300.308491.02267924.10115624.6647550.8855053641224.381483258670.000027.53397624.23140024.8211040.8519503641224.525066241320.000028.7869791.1004103.0459930.021909364122.072926-6.3311300.000094.615309130.508716133.5357180.90011736412132.015099258670.0000146.664441133.717705136.9636230.85670736412135.343844241320.0000157.981545
3301.1462503.0132310.022953364122.081317-6.0063630.000090.67705823.67263424.2268150.8859063641223.949209258450.000027.03357924.26330924.8543000.8517143641224.557815241990.000028.8334082.1907894.1384080.033151364123.162892-6.0063630.000095.407417129.202581132.2461090.88485836412130.716175258450.0000147.725544134.179053137.4258340.85869736412135.795209241990.0000158.140916
337-0.6654981.2000910.002967364120.269062-6.1742200.572990.68746623.47514924.0335610.8724123641223.752620252530.000027.22638423.53454024.1263720.8279623641223.831297235610.000028.7830960.3847372.3237140.014247364121.348729-6.1742200.006194.664781127.549702130.5817850.88007136412129.062425252530.0000146.649995129.611536132.8639200.83224636412131.228241235610.0000157.679686
3441.8660443.7351990.030736364122.800551-4.2329170.000091.11495922.39396622.9522000.8341813641222.673676244160.000027.18077223.19870223.7876020.8175313641223.492597232560.000028.7360202.9913034.9240290.042242364123.961977-4.2329170.000093.793202121.579390124.5710660.84411136412123.069395244160.0000145.797641127.830699131.0443550.82346536412129.431613232560.0000157.179219
351-1.8588710.0082530.01017536412-0.925162-6.0392050.052090.92752122.73563223.3085130.8257183641223.022386242740.000027.88164522.45686823.0485570.7925013641222.751856226190.000028.708948-0.3995301.5237820.006084364120.567723-6.0392050.247193.309607123.137885126.1900720.84418736412124.664052242740.0000147.673500123.953709127.2030930.79672736412125.575713226190.0000157.614568
358-10.321498-8.4609760.10388736412-9.387045-9.8780230.000090.35779225.60256626.2003220.8924253641225.901071263340.000029.02326222.86238723.4505230.8098883641223.156166236270.000028.591799-8.848608-6.9284910.08413736412-7.889462-9.8780230.000093.769321140.011507143.2155050.90492736412141.612132263340.0000156.490122127.335053130.5784860.81554536412128.943081236270.0000158.106636
\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "dfsdata = None\n", "if load_results:\n", " try:\n", " fname = path2 / 'dfsdata_{}_{}.pkl'.format(freq, country[1])\n", " with open(fname, 'rb') as file:\n", " dfsdata, nruns, c19idx = pickle.loads(file.read())\n", " print('{} loaded.\\n'.format(fname))\n", " except Exception as e:\n", " print(e, '\\nError in loading {}.\\nRunning analysis...'.format(fname))\n", " from bootstrap_stat import bootstrap_stat as bp\n", " \n", "if dfsdata is None:\n", " print('1/3')\n", " dfsdiff = describe(dfs['2020', freq],\n", " variable=['distance_d', 'duration_d'],\n", " new_variable=['distance', 'duration'], show=False)\n", " print('2/3')\n", " dfs2019 = describe(dfs['2019', freq],\n", " variable=['distance', 'duration'],\n", " new_variable=['distance2019', 'duration2019'], show=False)\n", " print('3/3')\n", " dfs2020 = describe(dfs['2020', freq],\n", " variable=['distance', 'duration'],\n", " new_variable=['distance2020', 'duration2020'], show=False)\n", " dfsdata = pd.concat([dfsdiff, dfs2019, dfs2020], axis=1)\n", " # add nathletes data\n", " for var in ['distance', 'duration']:\n", " dfsdata[var, 'nathletes'] = nathletesd[freq].values\n", " for year in ['2019', '2020']:\n", " dfsdata[var+year, 'nathletes'] = nathletes[year, freq].values\n", " dfsdata = dfsdata.sort_index(axis=1)\n", " # save data\n", " try:\n", " fname = path2 / 'dfsdata_{}_{}.pkl'.format(freq, country[1])\n", " with open(fname, 'wb') as file:\n", " pickle.dump((dfsdata, nruns, c19idx), file)\n", " except Exception as e:\n", " print(e, '\\nError in saving {}.'.format(fname))\n", "\n", "# display statistics\n", "title = 'Results for all athletes of country: {}'.format(country[0])\n", "display_dfsd(dfsdata, title)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Plots" ] }, { "cell_type": "code", "execution_count": 49, "metadata": { "ExecuteTime": { "end_time": "2022-03-12T04:14:39.084812Z", "start_time": "2022-03-12T04:14:37.839009Z" } }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "ds = 'default' # drawstyle\n", "ds = 'steps-post' # drawstyle\n", "weight = 'normal'\n", "days = np.r_[0, np.cumsum(dfs['2019', 'm'].index.unique().daysinmonth.values)[:]]\n", "if estimate == np.mean or estimate == np.nanmean:\n", " desc = 'mean'\n", "else:\n", " desc = '50%'\n", " sfrq = 'Daily'\n", "if freq == 'd':\n", " lw = 1\n", " factor = np.ones(int(days[-1]))\n", "elif freq == 'w':\n", " lw = 3\n", " factor = np.ones(int(days[-1]/7)) * 7\n", " sfrq = 'Weekly'\n", "elif freq == 'm':\n", " lw = 3 \n", " factor = np.diff(days)\n", " sfrq = 'Monthly'\n", " \n", "\n", "def repeat_last(df, freq):\n", " \"\"\"Repeat last point in data for plotting.\n", " \"\"\" \n", " x = df.iloc[-1]\n", " if freq == 'd':\n", " x.name = x.name + 1\n", " df = pd.concat([df, x.to_frame().transpose()])\n", " t = df.index\n", " elif freq == 'w':\n", " x.name = x.name + 7\n", " df = pd.concat([df, x.to_frame().transpose()])\n", " t = df.index\n", " elif freq == 'm':\n", " #x.name = x.name + 31\n", " x.name = days[-1]\n", " df = pd.concat([df, x.to_frame().transpose()])\n", " #t = df.index - 31\n", " t = days\n", " return t, df\n", "\n", "\n", "def plot(ax, t, df, c1, ds='default'):\n", " \"\"\"General plot.\n", " \"\"\"\n", " if ds == 'steps-post':\n", " ax.fill_between(t, df['ci_inf'], df['ci_sup'], label='', lw=lw+1,\n", " color=c1, alpha=.3, clip_on=False, step='post')\n", " else:\n", " ax.fill_between(t, df['ci_inf'], df['ci_sup'], label='', lw=lw+1,\n", " color=c1, alpha=.3, clip_on=False) \n", " ax2 = sns.lineplot(x=t, y=df['mean'], color=c1, ci=None, alpha=1, \n", " legend=False, lw=lw, ax=ax, clip_on=False, drawstyle=ds)\n", " return ax2\n", "\n", "\n", "def plot2(ax, axb, t, df, n, c1, c2, ds='default'):\n", " \"\"\"General plot.\n", " \"\"\"\n", " if ds == 'steps-post':\n", " ax.fill_between(t, df['ci_inf'], df['ci_sup'],\n", " color=c1, alpha=.4, clip_on=False, step='post')\n", " else:\n", " ax.fill_between(t, df['ci_inf'], df['ci_sup'],\n", " color=c1, alpha=.4, clip_on=False) \n", " sns.lineplot(x=t, y=df['mean'], color=c1, ci=None, alpha=1, \n", " legend=False, lw=lw+1, ax=ax, clip_on=False, drawstyle=ds)\n", " axb.plot(t, n, color=c2, lw=lw, alpha=1, clip_on=False, drawstyle=ds)\n", "\n", "\n", "for v, var in enumerate(variables):\n", " fig = plt.figure(figsize=(13, 8))\n", " gs = fig.add_gridspec(4, 1)\n", " ax0 = fig.add_subplot(gs[0, 0])\n", " ax1 = fig.add_subplot(gs[1, 0])\n", " ax2 = fig.add_subplot(gs[2:, 0])\n", "\n", " #plt.suptitle('{} long-distance running volume (all athletes from {})'.\n", " # format(sfrq, country[0]), fontsize=16, c=c3, ha='center', y=.98) \n", "\n", " for ax in [ax0, ax1, ax2]:\n", " ax.axvline(days[3], c=colorbase, alpha=.8, ls=':', lw=1, label='')\n", " ax.axvline(days[6], c=colorbase, alpha=.8, ls=':', lw=1, label='')\n", " ax.axvline(days[9], c=colorbase, alpha=.8, ls=':', lw=1, label='')\n", " ax.set_xlim([0, 365])\n", " ax.spines['left'].set_position(('axes', -0.012))\n", " \n", " t, var2019 = repeat_last(dfsdata[var+'2019'], freq)\n", " t, var2020 = repeat_last(dfsdata[var+'2020'], freq)\n", " \n", " # first plot\n", " ax0.plot(t, var2019['nathletes'], color=color2019, lw=lw, alpha=1, clip_on=False,\n", " drawstyle=ds, label='2019')\n", " ax0.plot(t, var2020['nathletes'], color=color2020, lw=lw, alpha=1, clip_on=False,\n", " drawstyle=ds, label='2020')\n", " ax0.text(0.01, 0.9, 'A', ha='right', weight=weight,\n", " fontsize=14, c='k', transform=ax0.transAxes)\n", " ax0.set_ylabel('Number of\\nathletes', c=colorbase, fontsize=16)\n", " handles, labels = ax0.get_legend_handles_labels()\n", " ax0.legend(handles, labels, frameon=False, loc='lower left')\n", " # second plot\n", " plot(ax1, t, var2019, color2019, ds)\n", " plot(ax1, t, var2020, color2020, ds)\n", " ax1.text(0.01, 0.9, 'B', ha='right', weight=weight,\n", " fontsize=14, c='k', transform=ax1.transAxes)\n", " ax1.set_ylabel('{} [{}]\\nMean and 95%CI'.format(var.capitalize(), units[v]),\n", " c=colorbase, fontsize=16)\n", " # third plot\n", " ax2b = ax2.twinx()\n", " ax2.axhline(0, c=colorbase, alpha=.8, lw=2)\n", " t, vard = repeat_last(dfsdata[var], freq)\n", " plot2(ax2, ax2b, t, vard, vard['nathletes'].values, colorvar, colorn, ds)\n", " ax2.text(0.01, 0.95, 'C', ha='right', weight=weight,\n", " fontsize=14, c='k', transform=ax2.transAxes)\n", " ax2.tick_params(axis='x', colors=colorbase)\n", " ax2.tick_params(axis='y', colors=colorvar)\n", " ax2.spines['left'].set_color(colorvar)\n", " ax2.spines['bottom'].set_position(('axes', -0.15))\n", " text = '{} difference [%]\\nMean and 95%CI'.format(var.capitalize())\n", " ax2.set_ylabel(text, color=colorvar, fontsize=16)\n", " ax2.tick_params(axis='both', which='major', labelsize=14)\n", " text = 'Number of athletes difference [%]'.format(sfrq)\n", " ax2b.set_ylabel(text, rotation=-90, color=colorn, fontsize=16)\n", " ax2b.yaxis.set_label_coords(1.08, .5)\n", " ax2b.tick_params(axis='both', which='major', labelsize=14)\n", " ax2b.spines['bottom'].set_visible(False)\n", " ax2b.spines['top'].set_visible(False)\n", " ax2b.spines['left'].set_visible(False)\n", " ax2b.spines['right'].set_color(colorn)\n", " ax2b.spines['right'].set_position(('axes', 1.01))\n", " ax2b.tick_params(axis='y', colors=colorn)\n", " ax2b.grid(False, axis='both')\n", " # plot coronavirus policy response index\n", " colori = colors[4]\n", " ax2c = ax2.twinx()\n", " c19idx.index = dfsdata.index\n", " t, c19idx2 = repeat_last(c19idx, freq)\n", " ax2c.plot(t, c19idx2['si'], color=colorind, alpha=.8, lw=lw+2, ls='-',\n", " clip_on=False, drawstyle=ds)\n", " ax2c.set_ylim([100, -100])\n", " ax2c.set_yticks([0, 50, 100]) \n", " ax2c.spines['bottom'].set_visible(False)\n", " ax2c.spines['top'].set_visible(False)\n", " ax2c.spines['left'].set_visible(False)\n", " ax2c.spines['right'].set_color(colorind)\n", " ax2c.spines['right'].set_position(('axes', 1.11))\n", " ax2c.tick_params(axis='both', which='major', labelsize=14)\n", " ax2c.set_ylabel('2020 stringency index', rotation=-90, color=colorind, fontsize=16)\n", " ax2c.tick_params(axis='y', colors=colorind)\n", " ax2c.yaxis.set_label_coords(1.17, .45)\n", " # pvalues plot\n", " pvalidx = np.where(dfsdata[var]['pvalue'] <= alphaS)[0] \n", " pval = (t[pvalidx] + factor[pvalidx]/2, ) \n", " ax2.eventplot(-np.ones_like(pval)*72, colors=colorvar, lineoffsets=pval,\n", " linelengths=np.atleast_2d(factor[pvalidx]), orientation='vertical',\n", " linewidth=7, alpha=0.4, clip_on=False)\n", " # effect sizes plot\n", " esi = np.where(dfsdata[var]['cohensd'] > 0.2)[0] \n", " es = (t[esi] + factor[esi]/2, ) \n", " ax2.eventplot(-np.ones_like(es)*68, colors=colorvar, lineoffsets=es,\n", " linelengths=np.atleast_2d(factor[esi]), orientation='vertical',\n", " linewidth=5, alpha=1, clip_on=False) \n", " \n", " if country[1] != 'all':\n", " ax0.text(1.1, .8, 'Number of athletes', ha='center', weight=weight,\n", " fontsize=14, c=colorbase, transform=ax0.transAxes)\n", " ax0.text(1.03, .6, '2019:', ha='left', weight=weight,\n", " fontsize=14, c=color2019, transform=ax0.transAxes)\n", " ax0.text(1.1, .6, '{:,}'.format(nruns['2019', 'd'].size),\n", " ha='left', weight=weight, fontsize=14, c=color2019, transform=ax0.transAxes)\n", " ax0.text(1.03, .4, '2020:', ha='left', weight=weight,\n", " fontsize=14, c=color2020, transform=ax0.transAxes)\n", " ax0.text(1.1, .4, '{:,}'.format(nruns['2020', 'd'].size),\n", " ha='left', weight=weight, fontsize=14, c=color2020, transform=ax0.transAxes)\n", " \n", " ax1.text(1.09, .8, 'Mean {}'.format(var), ha='center', weight=weight,\n", " fontsize=14, c=colorbase, transform=ax1.transAxes)\n", " ax1.text(1.03, .6, '2019:', ha='left', weight=weight,\n", " fontsize=14, c=color2019, transform=ax1.transAxes)\n", " ax1.text(1.1, .6, '{:.1f} {}'.format(\n", " np.round(dfs_stat['2019', freq][desc][var], 1), units[v]),\n", " ha='left', weight=weight, fontsize=14, c=color2019, transform=ax1.transAxes) \n", " ax1.text(1.03, .4, '2020:', ha='left', weight=weight,\n", " fontsize=14, c=color2020, transform=ax1.transAxes)\n", " ax1.text(1.1, .4, '{:.1f} {}'.format(\n", " np.round(dfs_stat['2020', freq][desc][var], 1), units[v]),\n", " ha='left', weight=weight, fontsize=14, c=color2020, transform=ax1.transAxes) \n", " ax2.text(-.01, -0.13, 'p value < 0.001', ha='right',\n", " weight=weight, fontsize=13, c=colorbase, transform=ax2.transAxes) \n", " ax2.text(-.01, -0.07, 'effect size > 0.2', ha='right',\n", " weight=weight, fontsize=13, c=colorbase, transform=ax2.transAxes)\n", " \n", " ax2c.grid(False, axis='both')\n", " if freq != 'd':\n", " ax2.set_ylim([-60, 60])\n", " ax2.set_yticks([-40, -20, 0, 20, 40])\n", " ax2b.set_ylim([-30, 30])\n", " ax2b.set_yticks([-20, -10, 0, 10, 20]) \n", " \n", " for ax in [ax0, ax1, ax2]:\n", " ax.xaxis.set_ticks_position('bottom')\n", " ax.yaxis.set_ticks_position('left')\n", " ax.set_xticklabels([])\n", " ax.set_xlabel('')\n", " ax.spines['top'].set_visible(False)\n", " ax.spines['right'].set_visible(False)\n", " ax.grid(False, axis='both')\n", " ax.tick_params(axis='both', which='major', labelsize=14)\n", " ax.set_xticks(days)\n", "\n", " labels = [' ' + m[0] for m in months]\n", " labels.append('')\n", " #ax0.set_xticklabels(labels, rotation=0, fontsize=14, ha='left')\n", " #ax1.set_xticklabels(labels, rotation=0, fontsize=14, ha='left')\n", " labels = [m[0:3] for m in months]\n", " labels.append('')\n", " ax2.set_xticklabels(labels, rotation=0, fontsize=14, ha='left') \n", " ax2.tick_params(axis='both', which='major', labelsize=14)\n", " labels = [m[0:3] for m in months]\n", " labels.append('')\n", " ax2.set_xlabel('Month of the year', c=colorbase, fontsize=16)\n", " ax2.xaxis.set_ticks_position('bottom')\n", " ax2.spines['bottom'].set_color(colorbase)\n", " ax2.set_xticklabels(labels, rotation=0, fontsize=14, ha='left')\n", " \n", " fig.align_ylabels([ax0, ax1, ax2]) \n", " ax0.margins(y=.2)\n", " ax1.margins(y=.2)\n", " \n", " plt.subplots_adjust(left=0.15, bottom=0.15, right=0.85, top=0.9, wspace=0.1, hspace=0.15)\n", "\n", " fig.savefig(path2 / 'figure{}_{}_{}_{}.png'.format(v+1, var, freq, country[1]), dpi=200)\n", " plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Figures. Comparison between 2019 and 2020 of long-distance running volume (distance and duration) and number of athletes. Blue color denotes the difference in athletes' weekly running volume (2020 - 2019) in each week of the year, normalized by the mean weekly running volume in 2019 (mean among all athletes; 95% confidence interval). Red color denotes the difference in the number of athletes running per week in each week of the year, normalized by the mean number of athletes running per week in 2019. The numbers in the upper left corner indicate the mean weekly running volume in 2019 and 2020 by athletes. In the upper right are the number of athletes running per week, in 2019 and 2020. At the bottom, the orange horizontal bars indicate a small effect size (between 0.2 and 0.5) and the light blue horizontal bars indicate a statistical difference (p<0.001) between the corresponding weeks of 2019 and 2020 for the weekly running volume. The green curve indicates the mean COVID-19 government response stringency index during 2020, weighted by the number of athletes in the country." ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "hide_input": false, "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.9.10" }, "nbTranslate": { "displayLangs": [ "*" ], "hotkey": "alt-t", "langInMainMenu": true, "sourceLang": "en", "targetLang": "fr", "useGoogleTranslate": true }, "toc": { "base_numbering": 1, "nav_menu": {}, "number_sections": true, "sideBar": true, "skip_h1_title": true, "title_cell": "Contents", "title_sidebar": "Contents", "toc_cell": true, "toc_position": {}, "toc_section_display": true, "toc_window_display": false }, "varInspector": { "cols": { "lenName": 16, "lenType": 16, "lenVar": 40 }, "kernels_config": { "python": { "delete_cmd_postfix": "", "delete_cmd_prefix": "del ", "library": "var_list.py", "varRefreshCmd": "print(var_dic_list())" }, "r": { "delete_cmd_postfix": ") ", "delete_cmd_prefix": "rm(", "library": "var_list.r", "varRefreshCmd": "cat(var_dic_list()) " } }, "position": { "height": "280px", "left": "793px", "right": "20px", "top": "121px", "width": "470px" }, "types_to_exclude": [ "module", "function", "builtin_function_or_method", "instance", "_Feature" ], "window_display": false } }, "nbformat": 4, "nbformat_minor": 4 }