{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Case2-1: DatetimeIndexを使ってみる \n", "Use DatetimeIndex" ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:17.664253", "start_time": "2016-09-17T13:51:16.221191" }, "collapsed": true }, "outputs": [], "source": [ "%matplotlib inline\n", "import seaborn as sns\n", "import datetime\n", "import numpy as np\n", "import pandas as pd" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:17.674194", "start_time": "2016-09-17T13:51:17.667192" }, "collapsed": true }, "outputs": [], "source": [ "# set display number of lines\n", "pd.options.display.max_rows = 10" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## pandas.date_rangeで日付範囲データを作成 \n", "Generate date range by pandas.date_range" ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:17.741220", "start_time": "2016-09-17T13:51:17.679191" }, "collapsed": false }, "outputs": [], "source": [ "date_range = pd.date_range('2015-1-1', '2016-12-31', freq='B')" ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:17.759194", "start_time": "2016-09-17T13:51:17.743203" }, "collapsed": false }, "outputs": [ { "data": { "text/plain": [ "DatetimeIndex(['2015-01-01', '2015-01-02', '2015-01-05', '2015-01-06',\n", " '2015-01-07', '2015-01-08', '2015-01-09', '2015-01-12',\n", " '2015-01-13', '2015-01-14',\n", " ...\n", " '2016-12-19', '2016-12-20', '2016-12-21', '2016-12-22',\n", " '2016-12-23', '2016-12-26', '2016-12-27', '2016-12-28',\n", " '2016-12-29', '2016-12-30'],\n", " dtype='datetime64[ns]', length=522, freq='B')" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "date_range" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### freqのパラメータ \n", "Frequency strings\n", "\n", "> DateOffset objects \n", "> http://pandas.pydata.org/pandas-docs/stable/timeseries.html#dateoffset-objects" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "|文字 |オフセットクラス |説明 |\n", "|:--------------------|:-----------------|:-----------------------------------------------------|\n", "|D |Day |暦通りの日次 |\n", "|B |BusinessDay |毎営業日 |\n", "|H |Hour |毎時 |\n", "|Tまたはmin |Minute |毎分 |\n", "|S |Second |毎秒 |\n", "|Lまたはms |Milli |毎ミリ秒 |\n", "|U |Micro |毎マイクロ秒 |\n", "|M |MonthEnd |暦通りの月末毎 |\n", "|BM |BusinessMonthEnd |月の最終営業日毎 |\n", "|MS |MonthBegin |暦通りの月初毎 |\n", "|BMS |BusinessMonthBegin|月の営業開始日毎 |\n", "|W-MON,W-TUE,... |Week |毎週指定した曜日毎 |\n", "|WOM-1MON,WOM-2MON,...|WeekOfMonth |月の1~4週目の指定した曜日毎 |\n", "|Q-JAN,Q-FEB,... |QuarterEnd |指定した月に年度が変わる前提で、四半期の暦通りの月末毎|\n", "|BQ-JAN,BQ-FEB,... |BusinessQuarterEnd|指定した月に年度が変わる前提で、四半期の最終営業日毎 |\n", "|QS-JAN,QS-FEB,... |QuarterBegin |指定した月に年度が変わる前提で、四半期の暦通りの月初毎|\n", "|BQS-JAN,BQS-FEB,... |QuarterBegin |指定した月に年度が変わる前提で、四半期の営業開始日毎 |\n", "|A-JAN,A-FEB,... |YearEnd |1年に1度、指定した月の暦通りの月末毎 |\n", "|BA-JAN,BA-FEB,... |BusinessYearEnd |1年に1度、指定した月の最終営業日毎 |\n", "|AS-JAN,AS-FEB,... |YearBegin |1年に1度、指定した月の暦通りの月初毎 |\n", "|BAS-JAN,BAS-FEB,... |YearBegin |1年に1度、指定した月の営業開始日毎 |" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "[Case1-3](Case1-3.ipynb)と同様に次の確率微分方程式にしたがう確率過程 Stを求める([幾何ブラウン運動](https://ja.wikipedia.org/wiki/幾何ブラウン運動 \"Wikipedia\")) \n", "A stochastic process St is said to follow a [GBM](https://en.wikipedia.org/wiki/Geometric_Brownian_motion) if it satisfies the following stochastic differential equation \n", "\n", "$$dS_t = \\mu S_t\\,dt + \\sigma S_t\\,dB_t$$ \n", "\n", "今回は日毎にデータを変化させていく \n", "Generate sample stock price per day" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:17.775195", "start_time": "2016-09-17T13:51:17.762195" }, "collapsed": true }, "outputs": [], "source": [ "def generate_stock_price(S0, t, r, sigma, path, steps):\n", " dt = float(t) / steps\n", " w = np.cumsum(np.reshape(np.random.standard_normal(path * steps), (path, steps)), 1) * (np.sqrt(dt))\n", " t = np.cumsum(np.ones((path, steps)), 1) * dt \n", " return S0 * np.exp((r - 0.5 * np.power(sigma, 2)) * t + sigma * w)" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:17.797188", "start_time": "2016-09-17T13:51:17.780192" }, "collapsed": false }, "outputs": [], "source": [ "days = len(date_range)\n", "stock_price = generate_stock_price(1000, days / 365.0, 0.001, 0.2, 3, days)\n", "df = pd.DataFrame(stock_price.T, index=date_range, columns=['price1', 'price2', 'price3'])" ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:17.826191", "start_time": "2016-09-17T13:51:17.802194" }, "collapsed": 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", "
price1price2price3
2015-01-01989.740629983.8832851013.366341
2015-01-02978.695980985.0320791012.314720
2015-01-05994.142925987.0584211035.949139
2015-01-06993.850903979.9889991030.669402
2015-01-07986.860088991.9713061042.060216
\n", "
" ], "text/plain": [ " price1 price2 price3\n", "2015-01-01 989.740629 983.883285 1013.366341\n", "2015-01-02 978.695980 985.032079 1012.314720\n", "2015-01-05 994.142925 987.058421 1035.949139\n", "2015-01-06 993.850903 979.988999 1030.669402\n", "2015-01-07 986.860088 991.971306 1042.060216" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.head()" ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:18.400188", "start_time": "2016-09-17T13:51:17.834195" }, "collapsed": false, "scrolled": false }, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAq0AAAHtCAYAAAAz7EdaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nhBAu7/8BUJAzybS573oAAAAASUVORK5CYII=\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "df.plot()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## datetime.date型で参照 \n", "select data in datetime.date" ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:18.420231", "start_time": "2016-09-17T13:51:18.403192" }, "collapsed": false }, "outputs": [ { "data": { "text/plain": [ "price1 923.276412\n", "price2 1005.628056\n", "price3 1120.888874\n", "Name: 2016-01-01 00:00:00, dtype: float64" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.loc[datetime.date(2016, 1, 1)]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## datetime.datetime型で参照 \n", "select data in datetime.date" ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:18.440194", "start_time": "2016-09-17T13:51:18.424191" }, "collapsed": false }, "outputs": [ { "data": { "text/plain": [ "price1 923.276412\n", "price2 1005.628056\n", "price3 1120.888874\n", "Name: 2016-01-01 00:00:00, dtype: float64" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.loc[datetime.datetime(2016, 1, 1, 0, 0)]" ] }, { "cell_type": "markdown", "metadata": { "collapsed": false }, "source": [ "## str型で参照 \n", "select data in str" ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:18.468203", "start_time": "2016-09-17T13:51:18.450192" }, "collapsed": false }, "outputs": [ { "data": { "text/plain": [ "price1 923.276412\n", "price2 1005.628056\n", "price3 1120.888874\n", "Name: 2016-01-01 00:00:00, dtype: float64" ] }, "execution_count": 11, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.loc['2016/1/1']" ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:18.487189", "start_time": "2016-09-17T13:51:18.471192" }, "collapsed": false }, "outputs": [ { "data": { "text/plain": [ "price1 923.276412\n", "price2 1005.628056\n", "price3 1120.888874\n", "Name: 2016-01-01 00:00:00, dtype: float64" ] }, "execution_count": 12, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.loc['20160101']" ] }, { "cell_type": "code", "execution_count": 13, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:18.514189", "start_time": "2016-09-17T13:51:18.491194" }, "collapsed": false }, "outputs": [ { "data": { "text/plain": [ "price1 923.276412\n", "price2 1005.628056\n", "price3 1120.888874\n", "Name: 2016-01-01 00:00:00, dtype: float64" ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.loc['Jan 01, 2016']" ] }, { "cell_type": "markdown", "metadata": { "collapsed": true }, "source": [ "## 年でスライス \n", "slice data by year" ] }, { "cell_type": "code", "execution_count": 14, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:18.554193", "start_time": "2016-09-17T13:51:18.518191" }, "collapsed": 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", "
price1price2price3
2016-01-01923.2764121005.6280561120.888874
2016-01-04940.646484998.0604681138.470680
2016-01-05933.050665987.7505451149.391673
2016-01-06938.822515989.1340291133.833442
2016-01-07933.242641981.3388731122.444465
\n", "
" ], "text/plain": [ " price1 price2 price3\n", "2016-01-01 923.276412 1005.628056 1120.888874\n", "2016-01-04 940.646484 998.060468 1138.470680\n", "2016-01-05 933.050665 987.750545 1149.391673\n", "2016-01-06 938.822515 989.134029 1133.833442\n", "2016-01-07 933.242641 981.338873 1122.444465" ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.loc['2016':].head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## 月でスライス \n", "slice data by month" ] }, { "cell_type": "code", "execution_count": 15, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:18.590189", "start_time": "2016-09-17T13:51:18.559202" }, "collapsed": 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", "
price1price2price3
2015-05-011057.487572912.6455881112.760240
2015-05-041052.539710911.4844851127.899948
2015-05-051065.361094913.5935381121.501779
2015-05-061064.154526917.1353571125.727790
2015-05-071049.255429922.5459731135.515160
\n", "
" ], "text/plain": [ " price1 price2 price3\n", "2015-05-01 1057.487572 912.645588 1112.760240\n", "2015-05-04 1052.539710 911.484485 1127.899948\n", "2015-05-05 1065.361094 913.593538 1121.501779\n", "2015-05-06 1064.154526 917.135357 1125.727790\n", "2015-05-07 1049.255429 922.545973 1135.515160" ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.loc['2015-5'].head()" ] }, { "cell_type": "code", "execution_count": 16, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:18.628189", "start_time": "2016-09-17T13:51:18.595192" }, "collapsed": 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", "
price1price2price3
2015-12-01901.023452904.5391041129.333413
2015-12-02895.479921895.7171681135.919600
2015-12-03906.573250905.5086231129.993463
2015-12-04910.097565910.0951761135.237073
2015-12-07903.014103926.7292171125.134069
............
2016-01-25884.219952978.7231191123.653912
2016-01-26896.079473962.9061201115.194525
2016-01-27899.893775956.3772751118.549986
2016-01-28902.807967966.6323431105.583831
2016-01-29909.536833959.2237241101.142623
\n", "

44 rows × 3 columns

\n", "
" ], "text/plain": [ " price1 price2 price3\n", "2015-12-01 901.023452 904.539104 1129.333413\n", "2015-12-02 895.479921 895.717168 1135.919600\n", "2015-12-03 906.573250 905.508623 1129.993463\n", "2015-12-04 910.097565 910.095176 1135.237073\n", "2015-12-07 903.014103 926.729217 1125.134069\n", "... ... ... ...\n", "2016-01-25 884.219952 978.723119 1123.653912\n", "2016-01-26 896.079473 962.906120 1115.194525\n", "2016-01-27 899.893775 956.377275 1118.549986\n", "2016-01-28 902.807967 966.632343 1105.583831\n", "2016-01-29 909.536833 959.223724 1101.142623\n", "\n", "[44 rows x 3 columns]" ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.loc['2015-12':'2016-01']" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## インデックスに含まれていないタイムスタンプを範囲に含めることもできる \n", "possible to include timestampe out of index" ] }, { "cell_type": "code", "execution_count": 17, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:18.660191", "start_time": "2016-09-17T13:51:18.631192" }, "collapsed": 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", "
price1price2price3
2016-01-04940.646484998.0604681138.470680
2016-01-05933.050665987.7505451149.391673
2016-01-06938.822515989.1340291133.833442
2016-01-07933.242641981.3388731122.444465
2016-01-08942.086665987.3000531124.170540
\n", "
" ], "text/plain": [ " price1 price2 price3\n", "2016-01-04 940.646484 998.060468 1138.470680\n", "2016-01-05 933.050665 987.750545 1149.391673\n", "2016-01-06 938.822515 989.134029 1133.833442\n", "2016-01-07 933.242641 981.338873 1122.444465\n", "2016-01-08 942.086665 987.300053 1124.170540" ] }, "execution_count": 17, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.loc['2016-01-02':'2016-01-10']" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## datetime.time型で参照 \n", "select data in datetime.time type" ] }, { "cell_type": "code", "execution_count": 18, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:18.678191", "start_time": "2016-09-17T13:51:18.663193" }, "collapsed": false }, "outputs": [], "source": [ "# Generate 100 hourly data\n", "ser = pd.Series(generate_stock_price(100, 0.2, 0, 0.2, 100, 1).T[0],\n", " index=pd.date_range('2016-01-01', periods=100, freq='H'))" ] }, { "cell_type": "code", "execution_count": 19, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:18.732193", "start_time": "2016-09-17T13:51:18.684207" }, "collapsed": false, "scrolled": false }, "outputs": [ { "data": { "text/plain": [ "2016-01-01 00:00:00 87.630594\n", "2016-01-01 01:00:00 94.592101\n", "2016-01-01 02:00:00 109.935888\n", "2016-01-01 03:00:00 113.025606\n", "2016-01-01 04:00:00 90.536927\n", "Freq: H, dtype: float64" ] }, "execution_count": 19, "metadata": {}, "output_type": "execute_result" } ], "source": [ "ser.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### datetime.timeで参照すると指定した時刻を抜き出すことができる \n", "Select specified time data by datetime.time" ] }, { "cell_type": "code", "execution_count": 20, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:18.773192", "start_time": "2016-09-17T13:51:18.742192" }, "collapsed": false }, "outputs": [ { "data": { "text/plain": [ "2016-01-01 09:00:00 111.702307\n", "2016-01-02 09:00:00 82.899518\n", "2016-01-03 09:00:00 103.516002\n", "2016-01-04 09:00:00 97.888412\n", "Freq: 24H, dtype: float64" ] }, "execution_count": 20, "metadata": {}, "output_type": "execute_result" } ], "source": [ "ser[datetime.time(9, 0)]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 内部では.at_time()メソッドが動作している \n", "In inside .at_time() is working" ] }, { "cell_type": "code", "execution_count": 21, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:18.816193", "start_time": "2016-09-17T13:51:18.779191" }, "collapsed": false }, "outputs": [ { "data": { "text/plain": [ "2016-01-01 09:00:00 111.702307\n", "2016-01-02 09:00:00 82.899518\n", "2016-01-03 09:00:00 103.516002\n", "2016-01-04 09:00:00 97.888412\n", "Freq: 24H, dtype: float64" ] }, "execution_count": 21, "metadata": {}, "output_type": "execute_result" } ], "source": [ "ser.at_time(datetime.time(9, 0))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### .between_time()メソッドを使うと、時刻の範囲を指定できる \n", "To specify time range using .between_time()" ] }, { "cell_type": "code", "execution_count": 22, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:18.863192", "start_time": "2016-09-17T13:51:18.824194" }, "collapsed": false }, "outputs": [ { "data": { "text/plain": [ "2016-01-01 09:00:00 111.702307\n", "2016-01-01 10:00:00 108.773511\n", "2016-01-01 11:00:00 107.222576\n", "2016-01-01 12:00:00 105.180062\n", "2016-01-02 09:00:00 82.899518\n", " ... \n", "2016-01-03 12:00:00 95.556950\n", "2016-01-04 09:00:00 97.888412\n", "2016-01-04 10:00:00 96.704581\n", "2016-01-04 11:00:00 120.614416\n", "2016-01-04 12:00:00 98.740763\n", "dtype: float64" ] }, "execution_count": 22, "metadata": {}, "output_type": "execute_result" } ], "source": [ "ser.between_time(datetime.time(9, 0), datetime.time(12, 0))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "---" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Case2-2: 四本値を作成、範囲を変換 \n", "Create OHLC data and covert time range" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## .resample()メソッドを使って月足に変換 \n", "Convert daily data into monthly data using .resample()" ] }, { "cell_type": "code", "execution_count": 23, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:18.907192", "start_time": "2016-09-17T13:51:18.866194" }, "collapsed": 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", "
openhighlowclose
2015-01-31989.7406291041.371901959.0802691041.371901
2015-02-281032.8922151053.8382091002.7338781041.320584
2015-03-311035.3192101056.0300271008.7169841041.224075
2015-04-301026.6018461067.1368141000.9300561057.973435
2015-05-311057.4875721065.361094962.684445962.684445
...............
2016-08-31944.4955701038.776056923.6724321038.776056
2016-09-301021.7968151021.796815975.163187975.163187
2016-10-31975.2513491032.668315975.2513491032.668315
2016-11-301030.2419841091.7905521009.0951861009.095186
2016-12-31994.3442281071.370257994.3442281058.799528
\n", "

24 rows × 4 columns

\n", "
" ], "text/plain": [ " open high low close\n", "2015-01-31 989.740629 1041.371901 959.080269 1041.371901\n", "2015-02-28 1032.892215 1053.838209 1002.733878 1041.320584\n", "2015-03-31 1035.319210 1056.030027 1008.716984 1041.224075\n", "2015-04-30 1026.601846 1067.136814 1000.930056 1057.973435\n", "2015-05-31 1057.487572 1065.361094 962.684445 962.684445\n", "... ... ... ... ...\n", "2016-08-31 944.495570 1038.776056 923.672432 1038.776056\n", "2016-09-30 1021.796815 1021.796815 975.163187 975.163187\n", "2016-10-31 975.251349 1032.668315 975.251349 1032.668315\n", "2016-11-30 1030.241984 1091.790552 1009.095186 1009.095186\n", "2016-12-31 994.344228 1071.370257 994.344228 1058.799528\n", "\n", "[24 rows x 4 columns]" ] }, "execution_count": 23, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df['price1'].resample('M').ohlc()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## 四本値から異なる頻度の四本値のデータに変換してみる \n", "convert an OHLC data into another OHLC data" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### [Yahoo Finance](https://finance.yahoo.com \"yahoo finance\")からS&P指数の四本値を取ってくる \n", "extract data from Yahoo Finance into a DataFrame" ] }, { "cell_type": "code", "execution_count": 24, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:19.872223", "start_time": "2016-09-17T13:51:18.912192" }, "collapsed": true }, "outputs": [], "source": [ "from pandas_datareader import data\n", "try:\n", " gspc = data.DataReader(\"^GSPC\", 'yahoo', datetime.datetime(2006, 1, 1), datetime.datetime(2015, 12, 31))\n", "except Exception:\n", " gspc = pd.read_msgpack('data/^GSPC.mpack')" ] }, { "cell_type": "code", "execution_count": 25, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:19.903192", "start_time": "2016-09-17T13:51:19.876192" }, "collapsed": 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", "
OpenHighLowCloseVolumeAdj Close
Date
2006-01-031248.2900391270.2199711245.7399901268.80004925545700001268.800049
2006-01-041268.8000491275.3699951267.7399901273.45996125153300001273.459961
2006-01-051273.4599611276.9100341270.3000491273.47998024333400001273.479980
2006-01-061273.4799801286.0899661273.4799801285.44995124465600001285.449951
2006-01-091285.4499511290.7800291284.8199461290.15002423014900001290.150024
\n", "
" ], "text/plain": [ " Open High Low Close Volume \\\n", "Date \n", "2006-01-03 1248.290039 1270.219971 1245.739990 1268.800049 2554570000 \n", "2006-01-04 1268.800049 1275.369995 1267.739990 1273.459961 2515330000 \n", "2006-01-05 1273.459961 1276.910034 1270.300049 1273.479980 2433340000 \n", "2006-01-06 1273.479980 1286.089966 1273.479980 1285.449951 2446560000 \n", "2006-01-09 1285.449951 1290.780029 1284.819946 1290.150024 2301490000 \n", "\n", " Adj Close \n", "Date \n", "2006-01-03 1268.800049 \n", "2006-01-04 1273.459961 \n", "2006-01-05 1273.479980 \n", "2006-01-06 1285.449951 \n", "2006-01-09 1290.150024 " ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" } ], "source": [ "gspc.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 普通にresample().ohlc()を実行すると・・・ \n", "Run resample().ohlc()" ] }, { "cell_type": "code", "execution_count": 26, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:20.075195", "start_time": "2016-09-17T13:51:19.907192" }, "collapsed": false, "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", "
OpenHighLow...CloseVolumeAdj Close
openhighlowcloseopenhighlowcloseopenhigh...lowcloseopenhighlowcloseopenhighlowclose
Date
2006-01-081248.2900391273.4799801248.2900391273.4799801270.2199711286.0899661270.2199711286.0899661245.7399901273.479980...1268.8000491285.44995125545700002554570000243334000024465600001268.8000491285.4499511268.8000491285.449951
2006-01-151285.4499511294.1800541285.4499511286.0600591290.7800291294.9000241288.9599611288.9599611284.8199461288.119995...1286.0600591287.60998523014900002406130000220651000022065100001290.1500241294.1800541286.0600591287.609985
2006-01-221287.6099851287.6099851277.9300541285.0400391287.6099851287.7900391282.9300541285.0400391278.6099851278.609985...1261.4899901261.48999021799700002845810000217997000028458100001282.9300541285.0400391261.4899901261.489990
2006-01-291261.4899901273.8299561261.4899901273.8299561268.1899411286.3800051268.1899411286.3800051261.4899901273.829956...1263.8199461283.71997122560700002856780000225607000026236200001263.8199461283.7199711263.8199461283.719971
2006-02-051283.7199711285.1999511270.8399661270.8399661287.9399411287.9399411270.8699951270.8699951283.5100101283.510010...1264.0300291264.03002922827300002708310000228221000022822100001285.1899411285.1899411264.0300291264.030029
\n", "

5 rows × 24 columns

\n", "
" ], "text/plain": [ " Open High \\\n", " open high low close open \n", "Date \n", "2006-01-08 1248.290039 1273.479980 1248.290039 1273.479980 1270.219971 \n", "2006-01-15 1285.449951 1294.180054 1285.449951 1286.060059 1290.780029 \n", "2006-01-22 1287.609985 1287.609985 1277.930054 1285.040039 1287.609985 \n", "2006-01-29 1261.489990 1273.829956 1261.489990 1273.829956 1268.189941 \n", "2006-02-05 1283.719971 1285.199951 1270.839966 1270.839966 1287.939941 \n", "\n", " Low \\\n", " high low close open high \n", "Date \n", "2006-01-08 1286.089966 1270.219971 1286.089966 1245.739990 1273.479980 \n", "2006-01-15 1294.900024 1288.959961 1288.959961 1284.819946 1288.119995 \n", "2006-01-22 1287.790039 1282.930054 1285.040039 1278.609985 1278.609985 \n", "2006-01-29 1286.380005 1268.189941 1286.380005 1261.489990 1273.829956 \n", "2006-02-05 1287.939941 1270.869995 1270.869995 1283.510010 1283.510010 \n", "\n", " ... Close Volume \\\n", " ... low close open high \n", "Date ... \n", "2006-01-08 ... 1268.800049 1285.449951 2554570000 2554570000 \n", "2006-01-15 ... 1286.060059 1287.609985 2301490000 2406130000 \n", "2006-01-22 ... 1261.489990 1261.489990 2179970000 2845810000 \n", "2006-01-29 ... 1263.819946 1283.719971 2256070000 2856780000 \n", "2006-02-05 ... 1264.030029 1264.030029 2282730000 2708310000 \n", "\n", " Adj Close \\\n", " low close open high low \n", "Date \n", "2006-01-08 2433340000 2446560000 1268.800049 1285.449951 1268.800049 \n", "2006-01-15 2206510000 2206510000 1290.150024 1294.180054 1286.060059 \n", "2006-01-22 2179970000 2845810000 1282.930054 1285.040039 1261.489990 \n", "2006-01-29 2256070000 2623620000 1263.819946 1283.719971 1263.819946 \n", "2006-02-05 2282210000 2282210000 1285.189941 1285.189941 1264.030029 \n", "\n", " \n", " close \n", "Date \n", "2006-01-08 1285.449951 \n", "2006-01-15 1287.609985 \n", "2006-01-22 1261.489990 \n", "2006-01-29 1283.719971 \n", "2006-02-05 1264.030029 \n", "\n", "[5 rows x 24 columns]" ] }, "execution_count": 26, "metadata": {}, "output_type": "execute_result" } ], "source": [ "gspc.resample('W').ohlc().head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 各列毎のohlcを集計してしまうので、四本値毎の集計した値から必要な値を持ってくる \n", "Select applicable data from above results" ] }, { "cell_type": "code", "execution_count": 27, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:20.183188", "start_time": "2016-09-17T13:51:20.079191" }, "collapsed": false }, "outputs": [], "source": [ "gspc_weekly_ohlc = gspc.resample('W').ohlc()[[('Open', 'open'), ('High', 'high'), ('Low', 'low'), ('Close', 'close')]]\n", "gspc_weekly_ohlc.columns = gspc_weekly_ohlc.columns.droplevel()" ] }, { "cell_type": "code", "execution_count": 28, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:20.210193", "start_time": "2016-09-17T13:51:20.186193" }, "collapsed": 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", "
openhighlowclose
Date
2006-01-081248.2900391286.0899661245.7399901285.449951
2006-01-151285.4499511294.9000241282.7800291287.609985
2006-01-221287.6099851287.7900391260.9200441261.489990
2006-01-291261.4899901286.3800051259.4200441283.719971
2006-02-051283.7199711287.9399411261.0200201264.030029
\n", "
" ], "text/plain": [ " open high low close\n", "Date \n", "2006-01-08 1248.290039 1286.089966 1245.739990 1285.449951\n", "2006-01-15 1285.449951 1294.900024 1282.780029 1287.609985\n", "2006-01-22 1287.609985 1287.790039 1260.920044 1261.489990\n", "2006-01-29 1261.489990 1286.380005 1259.420044 1283.719971\n", "2006-02-05 1283.719971 1287.939941 1261.020020 1264.030029" ] }, "execution_count": 28, "metadata": {}, "output_type": "execute_result" } ], "source": [ "gspc_weekly_ohlc.head()" ] }, { "cell_type": "markdown", "metadata": { "collapsed": true }, "source": [ "## 番外編: HighChartsでローソク足を表示してみる \n", "Appendix: visualize candle stick chart by HighCharts" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Python3の場合\n", "python-highchartsはPython3に対応していないので、forkしたリポジトリからインストール \n", "\n", "For Python3 \n", "Install python-highcharts from forked repository \n", "\n", " git clone -b issue37-Python3 https://github.com/JrtPec/python-highcharts.git\n", " cd python-highcharts\n", " python setup.py install" ] }, { "cell_type": "code", "execution_count": 29, "metadata": { "ExecuteTime": { "end_time": "2016-09-17T13:51:20.983193", "start_time": "2016-09-17T13:51:20.213190" }, "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Server running in the folder /notebooks at 127.0.0.1:42122\n" ] }, { "data": { "text/html": [ "\n", "\n", "\n", "\n", "
\n", "
\n", "
\n", " \n", "
\n", "
\n", " \n", "
\n", "
\n", "\n", "
\n", "
\n", "
\n", "

Adjust chart settings

\n", "
\n", "
\n", "
\n", "
\n", "
\n", "
\n", " \n", "
\n", "
\n", "
\n", "
\n", "
\n", " \n", "
.json
\n", "
\n", "
\n", " \n", "
\n", "\n", "
\n", "
\n", "
\n", "
\n", "
\n", "
\n", "\n", "
\n", "
\n", "
\n", "
\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n" ], "text/plain": [ "" ] }, "execution_count": 29, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import charts\n", "gspc['Time'] = gspc.index.astype(np.int64)//1000000\n", "series = [{'type':'candlestick',\n", " 'name':'SP500',\n", " 'data':gspc[['Time', 'Open', 'High', 'Low', 'Close']].values}]\n", "\n", "options = {'rangeSelector':{'selected':0},\n", " 'xAxis':{'gridLineWidth':1, 'gridLineDashStyle':'Dot'},\n", " 'yAxis':{'gridLineWidth':1, 'gridLineDashStyle':'Dot'}}\n", "\n", "charts.plot(series, stock=True, show='inline', options=options, display=['SP500'])" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": true }, "outputs": [], "source": [] } ], "metadata": { "hide_input": false, "kernelspec": { "display_name": "Python 3", "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": "3.5.1" }, "nav_menu": {}, "toc": { "navigate_menu": true, "number_sections": true, "sideBar": true, "threshold": 6, "toc_cell": false, "toc_section_display": "block", "toc_window_display": false } }, "nbformat": 4, "nbformat_minor": 1 }