{ "cells": [ { "cell_type": "markdown", "metadata": { "id": "view-in-github", "colab_type": "text" }, "source": [ "\"Open" ] }, { "cell_type": "markdown", "source": [ "Source of data: FactSets (All ETFs with some cleaning)" ], "metadata": { "id": "vpaUJCDDSrKd" } }, { "cell_type": "code", "execution_count": 182, "metadata": { "id": "2aCXQmsPr0QO" }, "outputs": [], "source": [ "import numpy as np\n", "import pandas as pd\n", "from pylab import mpl, plt\n", "import statsmodels.formula.api as smf\n", "import statsmodels.api as sm\n", "plt.style.use('seaborn')\n", "mpl.rcParams['font.family'] = 'serif'\n", "%matplotlib inline" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "id": "O3aELhCd_v3v", "outputId": "a2577fb0-5975-4904-b74e-449bc37c9ede", "colab": { "base_uri": "https://localhost:8080/", "height": 35 } }, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ "'/content'" ], "application/vnd.google.colaboratory.intrinsic+json": { "type": "string" } }, "metadata": {}, "execution_count": 2 } ], "source": [ "import os \n", "os.getcwd()" ] }, { "cell_type": "code", "execution_count": 120, "metadata": { "id": "9wQvtRuTr0QP" }, "outputs": [], "source": [ "filename = 'https://raw.githubusercontent.com/cyrus723/my-first-binder/main/factsets_etf.csv' " ] }, { "cell_type": "code", "execution_count": 121, "metadata": { "id": "9lVTFCFBr0QR", "uuid": "53a33e39-a3ff-4c95-b0f2-a94d727ae0da" }, "outputs": [], "source": [ "data = pd.read_csv(filename) " ] }, { "cell_type": "code", "execution_count": 122, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "ereRgc3er0QR", "outputId": "9004baf5-21e1-40c6-e063-9d0e2649e215", "uuid": "53a33e39-a3ff-4c95-b0f2-a94d727ae0da" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "\n", "RangeIndex: 1893 entries, 0 to 1892\n", "Data columns (total 8 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", " 0 ticker 1893 non-null object \n", " 1 name 1893 non-null object \n", " 2 issuer 1893 non-null object \n", " 3 exp 1893 non-null float64\n", " 4 nav 1893 non-null float64\n", " 5 aum 1893 non-null float64\n", " 6 class 1893 non-null object \n", " 7 beta 1893 non-null float64\n", "dtypes: float64(4), object(4)\n", "memory usage: 118.4+ KB\n" ] } ], "source": [ "data.info() " ] }, { "cell_type": "code", "execution_count": 123, "metadata": { "id": "O_0MH7A7_v3z" }, "outputs": [], "source": [ "data['scaled_aum'] = np.log(data['aum'])" ] }, { "cell_type": "code", "execution_count": 124, "metadata": { "id": "wcAFYk-T_v3z" }, "outputs": [], "source": [ "data['aum_million'] = data['aum'] / 1000000" ] }, { "cell_type": "code", "execution_count": 125, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 206 }, "id": "kPWQcRbsr0QR", "outputId": "d11a5f1d-8686-40de-e71a-fc60f21312c2" }, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ " ticker name \\\n", "0 SINV ETFMG Prime 2X Daily Inverse Junior Silver Min... \n", "1 B621WX WisdomTree Short AUD Long EUR \n", "2 B68GS4 WisdomTree Long EUR Short USD \n", "3 *CINC.B CI DoubleLine Income USD Fund Unhedged \n", "4 B68GSP WisdomTree Long JPY Short USD \n", "\n", " issuer exp nav aum class beta scaled_aum \\\n", "0 ETFMG 1.0 11.9 115432.0 Equity -2.1 11.656437 \n", "1 WisdomTree 0.0 30.2 156868.0 Currency -0.2 11.963160 \n", "2 WisdomTree 0.0 29.8 166949.0 Currency 0.1 12.025444 \n", "3 CI Financial Corp. 1.0 17.7 176787.0 Fixed Income -0.3 12.082701 \n", "4 WisdomTree 0.0 29.0 203339.0 Currency -0.1 12.222630 \n", "\n", " aum_million \n", "0 0.115432 \n", "1 0.156868 \n", "2 0.166949 \n", "3 0.176787 \n", "4 0.203339 " ], "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", "
tickernameissuerexpnavaumclassbetascaled_aumaum_million
0SINVETFMG Prime 2X Daily Inverse Junior Silver Min...ETFMG1.011.9115432.0Equity-2.111.6564370.115432
1B621WXWisdomTree Short AUD Long EURWisdomTree0.030.2156868.0Currency-0.211.9631600.156868
2B68GS4WisdomTree Long EUR Short USDWisdomTree0.029.8166949.0Currency0.112.0254440.166949
3*CINC.BCI DoubleLine Income USD Fund UnhedgedCI Financial Corp.1.017.7176787.0Fixed Income-0.312.0827010.176787
4B68GSPWisdomTree Long JPY Short USDWisdomTree0.029.0203339.0Currency-0.112.2226300.203339
\n", "
\n", " \n", " \n", " \n", "\n", " \n", "
\n", "
\n", " " ] }, "metadata": {}, "execution_count": 125 } ], "source": [ "data.head() " ] }, { "cell_type": "code", "execution_count": 126, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 337 }, "id": "R5hJOfsYr0QS", "outputId": "178627a4-85f1-401c-bd4e-35e7d7634920" }, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ " ticker name \\\n", "1888 BDR7WS Hanwha ARIRANG S&P500 ETF(H) \n", "1889 BMJJF5 Hanwha Arirang SYNTH-MSCI Emerging Markets ETF(H) \n", "1890 BG8FBY Hanwha ARIRANG KTB 3Y Futures ETF \n", "1891 B7W7BH Hanwha ARIRANG Dividend ETF \n", "1892 B76Y8C HANWHA ARIRANG 200 ETF \n", "\n", " issuer exp nav aum \\\n", "1888 HANWHA LIFE INSURANCE Co., Ltd. 0.0 17574.9 4.420000e+11 \n", "1889 HANWHA LIFE INSURANCE Co., Ltd. 1.0 10002.4 7.400000e+11 \n", "1890 HANWHA LIFE INSURANCE Co., Ltd. 0.0 51743.0 7.540000e+11 \n", "1891 HANWHA LIFE INSURANCE Co., Ltd. 0.0 12555.7 2.030000e+12 \n", "1892 HANWHA LIFE INSURANCE Co., Ltd. 0.0 35505.7 6.610000e+12 \n", "\n", " class beta scaled_aum aum_million \n", "1888 Equity 0.7 26.814576 442000.0 \n", "1889 Equity 0.8 27.329916 740000.0 \n", "1890 Fixed Income 0.0 27.348658 754000.0 \n", "1891 Equity 0.8 28.339057 2030000.0 \n", "1892 Equity 1.0 29.519605 6610000.0 " ], "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", "
tickernameissuerexpnavaumclassbetascaled_aumaum_million
1888BDR7WSHanwha ARIRANG S&P500 ETF(H)HANWHA LIFE INSURANCE Co., Ltd.0.017574.94.420000e+11Equity0.726.814576442000.0
1889BMJJF5Hanwha Arirang SYNTH-MSCI Emerging Markets ETF(H)HANWHA LIFE INSURANCE Co., Ltd.1.010002.47.400000e+11Equity0.827.329916740000.0
1890BG8FBYHanwha ARIRANG KTB 3Y Futures ETFHANWHA LIFE INSURANCE Co., Ltd.0.051743.07.540000e+11Fixed Income0.027.348658754000.0
1891B7W7BHHanwha ARIRANG Dividend ETFHANWHA LIFE INSURANCE Co., Ltd.0.012555.72.030000e+12Equity0.828.3390572030000.0
1892B76Y8CHANWHA ARIRANG 200 ETFHANWHA LIFE INSURANCE Co., Ltd.0.035505.76.610000e+12Equity1.029.5196056610000.0
\n", "
\n", " \n", " \n", " \n", "\n", " \n", "
\n", "
\n", " " ] }, "metadata": {}, "execution_count": 126 } ], "source": [ "data.tail() " ] }, { "cell_type": "markdown", "metadata": { "id": "xXCZB_aNr0QT" }, "source": [ "### Summary Statistics" ] }, { "cell_type": "code", "execution_count": 127, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "3ngKC-ior0QU", "outputId": "b8d99645-5565-41bc-d70b-e5e9d1100769" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "\n", "RangeIndex: 1893 entries, 0 to 1892\n", "Data columns (total 10 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", " 0 ticker 1893 non-null object \n", " 1 name 1893 non-null object \n", " 2 issuer 1893 non-null object \n", " 3 exp 1893 non-null float64\n", " 4 nav 1893 non-null float64\n", " 5 aum 1893 non-null float64\n", " 6 class 1893 non-null object \n", " 7 beta 1893 non-null float64\n", " 8 scaled_aum 1893 non-null float64\n", " 9 aum_million 1893 non-null float64\n", "dtypes: float64(6), object(4)\n", "memory usage: 148.0+ KB\n" ] } ], "source": [ "data.info() " ] }, { "cell_type": "code", "execution_count": 128, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 300 }, "id": "8O_XLOsEr0QU", "outputId": "9ad3971e-6d3f-4da9-94e1-bb3b72bad904" }, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ " exp nav aum beta scaled_aum aum_million\n", "count 1893.00 1893.00 1.893000e+03 1893.00 1893.00 1893.00\n", "mean 0.55 994.99 8.829291e+09 0.66 18.42 8829.29\n", "std 0.96 7030.29 1.620810e+11 0.67 2.55 162081.01\n", "min 0.00 0.00 1.154320e+05 -4.90 11.66 0.12\n", "25% 0.00 20.30 1.878199e+07 0.30 16.75 18.78\n", "50% 0.00 29.50 8.541894e+07 0.70 18.26 85.42\n", "75% 1.00 54.70 4.387002e+08 1.00 19.90 438.70\n", "max 20.00 108546.70 6.610000e+12 4.40 29.52 6610000.00" ], "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", "
expnavaumbetascaled_aumaum_million
count1893.001893.001.893000e+031893.001893.001893.00
mean0.55994.998.829291e+090.6618.428829.29
std0.967030.291.620810e+110.672.55162081.01
min0.000.001.154320e+05-4.9011.660.12
25%0.0020.301.878199e+070.3016.7518.78
50%0.0029.508.541894e+070.7018.2685.42
75%1.0054.704.387002e+081.0019.90438.70
max20.00108546.706.610000e+124.4029.526610000.00
\n", "
\n", " \n", " \n", " \n", "\n", " \n", "
\n", "
\n", " " ] }, "metadata": {}, "execution_count": 128 } ], "source": [ "data.describe().round(2) " ] }, { "cell_type": "code", "execution_count": 129, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 268 }, "id": "UOy5nm2Cr0QV", "outputId": "c7535723-a354-44e9-8453-74cd8dbb549a" }, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ " ticker name \\\n", "min *ARB 15 FiCAS Active Crypto ETP \n", "mean NaN NaN \n", "std NaN NaN \n", "median NaN NaN \n", "max ZIG iShares eb.rexx Government Germany UCITS ETF (DE) \n", "\n", " issuer exp nav aum \\\n", "min 21Shares AG 0.00 0.00 1.154320e+05 \n", "mean NaN 0.55 994.99 8.829291e+09 \n", "std NaN 0.96 7030.29 1.620810e+11 \n", "median NaN 0.00 29.50 8.541894e+07 \n", "max Zacks Investment Management 20.00 108546.70 6.610000e+12 \n", "\n", " class beta scaled_aum aum_million \n", "min Alternatives -4.90 11.66 0.12 \n", "mean NaN 0.66 18.42 8829.29 \n", "std NaN 0.67 2.55 162081.01 \n", "median NaN 0.70 18.26 85.42 \n", "max Fixed Income 4.40 29.52 6610000.00 " ], "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", "
tickernameissuerexpnavaumclassbetascaled_aumaum_million
min*ARB15 FiCAS Active Crypto ETP21Shares AG0.000.001.154320e+05Alternatives-4.9011.660.12
meanNaNNaNNaN0.55994.998.829291e+09NaN0.6618.428829.29
stdNaNNaNNaN0.967030.291.620810e+11NaN0.672.55162081.01
medianNaNNaNNaN0.0029.508.541894e+07NaN0.7018.2685.42
maxZIGiShares eb.rexx Government Germany UCITS ETF (DE)Zacks Investment Management20.00108546.706.610000e+12Fixed Income4.4029.526610000.00
\n", "
\n", " \n", " \n", " \n", "\n", " \n", "
\n", "
\n", " " ] }, "metadata": {}, "execution_count": 129 } ], "source": [ "data.aggregate([min, \n", " np.mean, \n", " np.std, \n", " np.median, \n", " max] \n", ").round(2)" ] }, { "cell_type": "code", "source": [ "data.info()" ], "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "FKZTXgjvYFvn", "outputId": "a3727edb-e084-4c76-c1c6-32b34e46e338" }, "execution_count": 131, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "\n", "RangeIndex: 1893 entries, 0 to 1892\n", "Data columns (total 10 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", " 0 ticker 1893 non-null object \n", " 1 name 1893 non-null object \n", " 2 issuer 1893 non-null object \n", " 3 exp 1893 non-null float64\n", " 4 nav 1893 non-null float64\n", " 5 aum 1893 non-null float64\n", " 6 class 1893 non-null object \n", " 7 beta 1893 non-null float64\n", " 8 scaled_aum 1893 non-null float64\n", " 9 aum_million 1893 non-null float64\n", "dtypes: float64(6), object(4)\n", "memory usage: 148.0+ KB\n" ] } ] }, { "cell_type": "code", "execution_count": 132, "metadata": { "id": "X_re1aq1_v32", "outputId": "4baa92d3-3359-4fcd-eed5-37c95a987ac9", "colab": { "base_uri": "https://localhost:8080/", "height": 269 } }, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ " exp nav aum beta scaled_aum \\\n", "class \n", "Alternatives 1.820513 21.320513 1.186542e+08 0.235897 17.267977 \n", "Asset Allocation 0.965517 395.551724 3.299403e+09 0.544828 17.414796 \n", "Commodities 0.412162 175.028378 1.406693e+09 0.477027 17.345794 \n", "Currency 1.032967 758.836264 2.028511e+08 0.648352 16.333805 \n", "Equity 0.569337 852.437812 1.261168e+10 0.868906 18.660249 \n", "Fixed Income 0.257576 1957.354293 4.164261e+09 0.168434 18.875313 \n", "\n", " aum_million \n", "class \n", "Alternatives 118.654158 \n", "Asset Allocation 3299.403427 \n", "Commodities 1406.693136 \n", "Currency 202.851121 \n", "Equity 12611.676815 \n", "Fixed Income 4164.260947 " ], "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", "
expnavaumbetascaled_aumaum_million
class
Alternatives1.82051321.3205131.186542e+080.23589717.267977118.654158
Asset Allocation0.965517395.5517243.299403e+090.54482817.4147963299.403427
Commodities0.412162175.0283781.406693e+090.47702717.3457941406.693136
Currency1.032967758.8362642.028511e+080.64835216.333805202.851121
Equity0.569337852.4378121.261168e+100.86890618.66024912611.676815
Fixed Income0.2575761957.3542934.164261e+090.16843418.8753134164.260947
\n", "
\n", " \n", " \n", " \n", "\n", " \n", "
\n", "
\n", " " ] }, "metadata": {}, "execution_count": 132 } ], "source": [ "data2 = data.groupby(by='class').mean()\n", "data2" ] }, { "cell_type": "code", "source": [ "data['class'] = data['class'].replace(['Asset Allocation','Fixed Income' ], ['Asset_Allocation', \"Fixed_Income\"])" ], "metadata": { "id": "qPQlXv42cN2O" }, "execution_count": 138, "outputs": [] }, { "cell_type": "code", "source": [ "data2 = data.groupby(by='class').mean()\n", "data2" ], "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 269 }, "id": "YKs4Lig8arR2", "outputId": "65fb9dc8-84ff-44bc-b028-62d385d0b768" }, "execution_count": 143, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ " exp nav aum beta scaled_aum \\\n", "class \n", "Alternatives 1.820513 21.320513 1.186542e+08 0.235897 17.267977 \n", "Asset_Allocation 0.965517 395.551724 3.299403e+09 0.544828 17.414796 \n", "Commodities 0.412162 175.028378 1.406693e+09 0.477027 17.345794 \n", "Currency 1.032967 758.836264 2.028511e+08 0.648352 16.333805 \n", "Equity 0.569337 852.437812 1.261168e+10 0.868906 18.660249 \n", "Fixed_Income 0.257576 1957.354293 4.164261e+09 0.168434 18.875313 \n", "\n", " aum_million \n", "class \n", "Alternatives 118.654158 \n", "Asset_Allocation 3299.403427 \n", "Commodities 1406.693136 \n", "Currency 202.851121 \n", "Equity 12611.676815 \n", "Fixed_Income 4164.260947 " ], "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", "
expnavaumbetascaled_aumaum_million
class
Alternatives1.82051321.3205131.186542e+080.23589717.267977118.654158
Asset_Allocation0.965517395.5517243.299403e+090.54482817.4147963299.403427
Commodities0.412162175.0283781.406693e+090.47702717.3457941406.693136
Currency1.032967758.8362642.028511e+080.64835216.333805202.851121
Equity0.569337852.4378121.261168e+100.86890618.66024912611.676815
Fixed_Income0.2575761957.3542934.164261e+090.16843418.8753134164.260947
\n", "
\n", " \n", " \n", " \n", "\n", " \n", "
\n", "
\n", " " ] }, "metadata": {}, "execution_count": 143 } ] }, { "cell_type": "code", "source": [ "data3 = data.rename({'class': 'asset_class'}, axis=1, inplace=False)\n" ], "metadata": { "id": "snyOGy4LTklC" }, "execution_count": 144, "outputs": [] }, { "cell_type": "code", "source": [ "data.info()" ], "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "o7EXMCRdTkhQ", "outputId": "e02b564a-0eb1-49bc-fc2a-86aedc646b83" }, "execution_count": 145, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "\n", "RangeIndex: 1893 entries, 0 to 1892\n", "Data columns (total 10 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", " 0 ticker 1893 non-null object \n", " 1 name 1893 non-null object \n", " 2 issuer 1893 non-null object \n", " 3 exp 1893 non-null float64\n", " 4 nav 1893 non-null float64\n", " 5 aum 1893 non-null float64\n", " 6 class 1893 non-null object \n", " 7 beta 1893 non-null float64\n", " 8 scaled_aum 1893 non-null float64\n", " 9 aum_million 1893 non-null float64\n", "dtypes: float64(6), object(4)\n", "memory usage: 148.0+ KB\n" ] } ] }, { "cell_type": "code", "source": [ "data3.info()" ], "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "YkI8rx5QbJUX", "outputId": "6322e6a0-7383-4ab2-be76-cf22f50fc9e0" }, "execution_count": 146, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "\n", "RangeIndex: 1893 entries, 0 to 1892\n", "Data columns (total 10 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", " 0 ticker 1893 non-null object \n", " 1 name 1893 non-null object \n", " 2 issuer 1893 non-null object \n", " 3 exp 1893 non-null float64\n", " 4 nav 1893 non-null float64\n", " 5 aum 1893 non-null float64\n", " 6 asset_class 1893 non-null object \n", " 7 beta 1893 non-null float64\n", " 8 scaled_aum 1893 non-null float64\n", " 9 aum_million 1893 non-null float64\n", "dtypes: float64(6), object(4)\n", "memory usage: 148.0+ KB\n" ] } ] }, { "cell_type": "code", "source": [ "data4 = pd.get_dummies(data3, columns=['asset_class'])" ], "metadata": { "id": "LYigW3UuTkey" }, "execution_count": 149, "outputs": [] }, { "cell_type": "code", "source": [ "data4.info()" ], "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "sZeGtM2dZD40", "outputId": "b5456cb9-cf42-498e-bf10-a263d2e0424c" }, "execution_count": 150, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "\n", "RangeIndex: 1893 entries, 0 to 1892\n", "Data columns (total 15 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", " 0 ticker 1893 non-null object \n", " 1 name 1893 non-null object \n", " 2 issuer 1893 non-null object \n", " 3 exp 1893 non-null float64\n", " 4 nav 1893 non-null float64\n", " 5 aum 1893 non-null float64\n", " 6 beta 1893 non-null float64\n", " 7 scaled_aum 1893 non-null float64\n", " 8 aum_million 1893 non-null float64\n", " 9 asset_class_Alternatives 1893 non-null uint8 \n", " 10 asset_class_Asset_Allocation 1893 non-null uint8 \n", " 11 asset_class_Commodities 1893 non-null uint8 \n", " 12 asset_class_Currency 1893 non-null uint8 \n", " 13 asset_class_Equity 1893 non-null uint8 \n", " 14 asset_class_Fixed_Income 1893 non-null uint8 \n", "dtypes: float64(6), object(3), uint8(6)\n", "memory usage: 144.3+ KB\n" ] } ] }, { "cell_type": "markdown", "metadata": { "id": "M6xOtd9Mr0Qi" }, "source": [ "## Regression Analysis" ] }, { "cell_type": "code", "execution_count": 157, "metadata": { "id": "FUe1C6Tlr0Qi", "uuid": "0bc55f0d-cd99-45b9-955e-3a126360e94f" }, "outputs": [], "source": [ "data5 = data4[['exp', \n", " 'scaled_aum', \n", " 'beta', \n", " 'asset_class_Alternatives',\n", " 'asset_class_Commodities',\n", " 'asset_class_Currency',\n", " 'asset_class_Equity',\n", " 'asset_class_Fixed_Income']].dropna()" ] }, { "cell_type": "code", "execution_count": 158, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 364 }, "id": "lcttj8upr0Qi", "outputId": "ccc502df-2a04-4e34-e725-42cff865ba5a", "scrolled": true, "uuid": "73526ac3-4bf0-4455-89b2-f6aa614ffdca" }, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ " exp scaled_aum beta asset_class_Alternatives \\\n", "count 1893.000000 1893.000000 1893.000000 1893.000000 \n", "mean 0.552034 18.423791 0.658162 0.020602 \n", "std 0.964520 2.549140 0.665928 0.142086 \n", "min 0.000000 11.656437 -4.900000 0.000000 \n", "25% 0.000000 16.748409 0.300000 0.000000 \n", "50% 0.000000 18.263078 0.700000 0.000000 \n", "75% 1.000000 19.899327 1.000000 0.000000 \n", "max 20.000000 29.519605 4.400000 1.000000 \n", "\n", " asset_class_Commodities asset_class_Currency asset_class_Equity \\\n", "count 1893.000000 1893.000000 1893.000000 \n", "mean 0.078183 0.048072 0.613312 \n", "std 0.268530 0.213975 0.487120 \n", "min 0.000000 0.000000 0.000000 \n", "25% 0.000000 0.000000 0.000000 \n", "50% 0.000000 0.000000 1.000000 \n", "75% 0.000000 0.000000 1.000000 \n", "max 1.000000 1.000000 1.000000 \n", "\n", " asset_class_Fixed_Income \n", "count 1893.000000 \n", "mean 0.209192 \n", "std 0.406839 \n", "min 0.000000 \n", "25% 0.000000 \n", "50% 0.000000 \n", "75% 0.000000 \n", "max 1.000000 " ], "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", "
expscaled_aumbetaasset_class_Alternativesasset_class_Commoditiesasset_class_Currencyasset_class_Equityasset_class_Fixed_Income
count1893.0000001893.0000001893.0000001893.0000001893.0000001893.0000001893.0000001893.000000
mean0.55203418.4237910.6581620.0206020.0781830.0480720.6133120.209192
std0.9645202.5491400.6659280.1420860.2685300.2139750.4871200.406839
min0.00000011.656437-4.9000000.0000000.0000000.0000000.0000000.000000
25%0.00000016.7484090.3000000.0000000.0000000.0000000.0000000.000000
50%0.00000018.2630780.7000000.0000000.0000000.0000001.0000000.000000
75%1.00000019.8993271.0000000.0000000.0000000.0000001.0000000.000000
max20.00000029.5196054.4000001.0000001.0000001.0000001.0000001.000000
\n", "
\n", " \n", " \n", " \n", "\n", " \n", "
\n", "
\n", " " ] }, "metadata": {}, "execution_count": 158 } ], "source": [ "data5.describe()" ] }, { "cell_type": "code", "execution_count": 159, "metadata": { "id": "_W9vzsfK_v33" }, "outputs": [], "source": [ "data6 = data5.loc[data5['exp'] <= 3]" ] }, { "cell_type": "code", "execution_count": 160, "metadata": { "id": "2kxVIQkt_v33", "outputId": "0fd54228-bb05-4176-c9c0-903f6fe069a6", "colab": { "base_uri": "https://localhost:8080/", "height": 364 } }, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ " exp scaled_aum beta asset_class_Alternatives \\\n", "count 1883.000000 1883.000000 1883.000000 1883.000000 \n", "mean 0.506638 18.426578 0.659480 0.018056 \n", "std 0.583454 2.550354 0.666672 0.133190 \n", "min 0.000000 11.656437 -4.900000 0.000000 \n", "25% 0.000000 16.748837 0.300000 0.000000 \n", "50% 0.000000 18.263078 0.700000 0.000000 \n", "75% 1.000000 19.901608 1.000000 0.000000 \n", "max 3.000000 29.519605 4.400000 1.000000 \n", "\n", " asset_class_Commodities asset_class_Currency asset_class_Equity \\\n", "count 1883.000000 1883.000000 1883.000000 \n", "mean 0.078598 0.047796 0.614976 \n", "std 0.269182 0.213391 0.486730 \n", "min 0.000000 0.000000 0.000000 \n", "25% 0.000000 0.000000 0.000000 \n", "50% 0.000000 0.000000 1.000000 \n", "75% 0.000000 0.000000 1.000000 \n", "max 1.000000 1.000000 1.000000 \n", "\n", " asset_class_Fixed_Income \n", "count 1883.000000 \n", "mean 0.210303 \n", "std 0.407632 \n", "min 0.000000 \n", "25% 0.000000 \n", "50% 0.000000 \n", "75% 0.000000 \n", "max 1.000000 " ], "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", "
expscaled_aumbetaasset_class_Alternativesasset_class_Commoditiesasset_class_Currencyasset_class_Equityasset_class_Fixed_Income
count1883.0000001883.0000001883.0000001883.0000001883.0000001883.0000001883.0000001883.000000
mean0.50663818.4265780.6594800.0180560.0785980.0477960.6149760.210303
std0.5834542.5503540.6666720.1331900.2691820.2133910.4867300.407632
min0.00000011.656437-4.9000000.0000000.0000000.0000000.0000000.000000
25%0.00000016.7488370.3000000.0000000.0000000.0000000.0000000.000000
50%0.00000018.2630780.7000000.0000000.0000000.0000001.0000000.000000
75%1.00000019.9016081.0000000.0000000.0000000.0000001.0000000.000000
max3.00000029.5196054.4000001.0000001.0000001.0000001.0000001.000000
\n", "
\n", " \n", " \n", " \n", "\n", " \n", "
\n", "
\n", " " ] }, "metadata": {}, "execution_count": 160 } ], "source": [ "data6.describe()" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "VDf11Dl0_v33" }, "outputs": [], "source": [ "" ] }, { "cell_type": "code", "execution_count": 161, "metadata": { "id": "7dKbV2Elr0Qj", "uuid": "07158c72-907f-4636-ad40-95182b7728e3" }, "outputs": [], "source": [ "data7 = data6[['exp', 'scaled_aum']].dropna()" ] }, { "cell_type": "code", "execution_count": 162, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 397 }, "id": "9wbJmB71r0Qk", "outputId": "b4d8e043-54d7-4356-e524-70bcd29e911b" }, "outputs": [ { "output_type": "display_data", "data": { "text/plain": [ "
" ], "image/png": "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\n" }, "metadata": { "needs_background": "light" } } ], "source": [ "pd.plotting.scatter_matrix(data7, \n", " alpha=0.2, \n", " diagonal='hist', \n", " hist_kwds={'bins': 35}, \n", " figsize=(10, 6));\n", "# plt.savefig('../../images/ch08/fts_11.png');" ] }, { "cell_type": "code", "execution_count": 163, "metadata": { "id": "xNTiylHo_v34" }, "outputs": [], "source": [ "data7 = data6[['exp', 'beta']].dropna()" ] }, { "cell_type": "code", "execution_count": 164, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 391 }, "outputId": "3d9c2711-db75-417f-8c67-60ed1835dc0e", "id": "1iIrgFqP_v34" }, "outputs": [ { "output_type": "display_data", "data": { "text/plain": [ "
" ], "image/png": "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\n" }, "metadata": { "needs_background": "light" } } ], "source": [ "pd.plotting.scatter_matrix(data7, \n", " alpha=0.2, \n", " diagonal='hist', \n", " hist_kwds={'bins': 35}, \n", " figsize=(10, 6));\n", "# plt.savefig('../../images/ch08/fts_11.png');" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "BtzeHjWO_v34" }, "outputs": [], "source": [ "" ] }, { "cell_type": "markdown", "metadata": { "id": "979Bgqb3r0Ql" }, "source": [ "### OLS Regression" ] }, { "cell_type": "code", "execution_count": 165, "metadata": { "id": "YTl2_rder0Ql" }, "outputs": [], "source": [ "reg = np.polyfit(data6['exp'], data6['scaled_aum'], deg=1) " ] }, { "cell_type": "code", "execution_count": 166, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 388 }, "id": "mG75szbyr0Ql", "outputId": "58e641cf-9cd7-480c-9a9b-786df06bf600", "uuid": "24c708df-1e81-48c6-b1c2-890dd52e541f" }, "outputs": [ { "output_type": "display_data", "data": { "text/plain": [ "
" ], "image/png": "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\n" }, "metadata": { "needs_background": "light" } } ], "source": [ "ax = data6.plot(kind='scatter', x='scaled_aum', y='exp', figsize=(10, 6)) \n", "ax.plot(data6['scaled_aum'], np.polyval(reg, data6['scaled_aum']), 'r', lw=2); \n", "# plt.savefig('../../images/ch08/fts_12.png');" ] }, { "cell_type": "code", "execution_count": 183, "metadata": { "id": "nSYG-6MV_v35", "outputId": "812cdc9e-d23a-4638-bd3f-ed889e8a5a88", "colab": { "base_uri": "https://localhost:8080/" } }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ " OLS Regression Results \n", "==============================================================================\n", "Dep. Variable: exp R-squared: 0.203\n", "Model: OLS Adj. R-squared: 0.201\n", "Method: Least Squares F-statistic: 79.80\n", "Date: Wed, 18 May 2022 Prob (F-statistic): 4.58e-89\n", "Time: 23:41:37 Log-Likelihood: -1442.8\n", "No. Observations: 1883 AIC: 2900.\n", "Df Residuals: 1876 BIC: 2938.\n", "Df Model: 6 \n", "Covariance Type: nonrobust \n", "============================================================================================\n", " coef std err t P>|t| [0.025 0.975]\n", "--------------------------------------------------------------------------------------------\n", "Intercept 2.2168 0.101 21.937 0.000 2.019 2.415\n", "scaled_aum -0.0697 0.005 -14.250 0.000 -0.079 -0.060\n", "beta 0.0669 0.020 3.335 0.001 0.028 0.106\n", "asset_class_Commodities -0.6271 0.069 -9.024 0.000 -0.763 -0.491\n", "asset_class_Currency -0.1260 0.078 -1.618 0.106 -0.279 0.027\n", "asset_class_Equity -0.4496 0.058 -7.785 0.000 -0.563 -0.336\n", "asset_class_Fixed_Income -0.6544 0.061 -10.670 0.000 -0.775 -0.534\n", "==============================================================================\n", "Omnibus: 210.403 Durbin-Watson: 1.932\n", "Prob(Omnibus): 0.000 Jarque-Bera (JB): 363.791\n", "Skew: 0.752 Prob(JB): 1.01e-79\n", "Kurtosis: 4.541 Cond. No. 207.\n", "==============================================================================\n", "\n", "Warnings:\n", "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n" ] } ], "source": [ "formula = 'exp ~ scaled_aum + beta + asset_class_Commodities+\tasset_class_Currency+\tasset_class_Equity\t+asset_class_Fixed_Income'\n", "results = smf.ols(formula, data6).fit()\n", "print(results.summary())\n" ] }, { "cell_type": "markdown", "metadata": { "id": "jassm1tFr0Ql" }, "source": [ "### Correlation" ] }, { "cell_type": "code", "execution_count": 171, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 238 }, "id": "KRjnhpywr0Qm", "outputId": "2dd814fb-ca11-405a-dd3e-6277b87f4b51", "uuid": "e1f9009e-5b73-4e04-9b10-deea36f4e508" }, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ " exp nav aum beta scaled_aum aum_million\n", "exp 1.000000 -0.626718 -0.445960 -0.164197 -0.559892 -0.445960\n", "nav -0.626718 1.000000 0.335254 -0.191956 0.633204 0.335254\n", "aum -0.445960 0.335254 1.000000 0.587259 0.730016 1.000000\n", "beta -0.164197 -0.191956 0.587259 1.000000 -0.105800 0.587259\n", "scaled_aum -0.559892 0.633204 0.730016 -0.105800 1.000000 0.730016\n", "aum_million -0.445960 0.335254 1.000000 0.587259 0.730016 1.000000" ], "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", "
expnavaumbetascaled_aumaum_million
exp1.000000-0.626718-0.445960-0.164197-0.559892-0.445960
nav-0.6267181.0000000.335254-0.1919560.6332040.335254
aum-0.4459600.3352541.0000000.5872590.7300161.000000
beta-0.164197-0.1919560.5872591.000000-0.1058000.587259
scaled_aum-0.5598920.6332040.730016-0.1058001.0000000.730016
aum_million-0.4459600.3352541.0000000.5872590.7300161.000000
\n", "
\n", " \n", " \n", " \n", "\n", " \n", "
\n", "
\n", " " ] }, "metadata": {}, "execution_count": 171 } ], "source": [ "data2.corr() " ] }, { "cell_type": "code", "execution_count": 179, "metadata": { "id": "RPdZ6TGU_v36" }, "outputs": [], "source": [ "X = data6[['scaled_aum', 'beta','asset_class_Commodities','asset_class_Currency','asset_class_Equity','asset_class_Fixed_Income']]\n", "y = data6['exp']" ] }, { "cell_type": "code", "source": [ "X_ols = sm.add_constant(X)\n", "model = sm.OLS(y, X_ols).fit()\n", "print(model.summary())" ], "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "fZxNlqGTiutk", "outputId": "c93b7944-067b-4bc5-95f7-bdc96aaba945" }, "execution_count": 184, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ " OLS Regression Results \n", "==============================================================================\n", "Dep. Variable: exp R-squared: 0.203\n", "Model: OLS Adj. R-squared: 0.201\n", "Method: Least Squares F-statistic: 79.80\n", "Date: Wed, 18 May 2022 Prob (F-statistic): 4.58e-89\n", "Time: 23:41:41 Log-Likelihood: -1442.8\n", "No. Observations: 1883 AIC: 2900.\n", "Df Residuals: 1876 BIC: 2938.\n", "Df Model: 6 \n", "Covariance Type: nonrobust \n", "============================================================================================\n", " coef std err t P>|t| [0.025 0.975]\n", "--------------------------------------------------------------------------------------------\n", "const 2.2168 0.101 21.937 0.000 2.019 2.415\n", "scaled_aum -0.0697 0.005 -14.250 0.000 -0.079 -0.060\n", "beta 0.0669 0.020 3.335 0.001 0.028 0.106\n", "asset_class_Commodities -0.6271 0.069 -9.024 0.000 -0.763 -0.491\n", "asset_class_Currency -0.1260 0.078 -1.618 0.106 -0.279 0.027\n", "asset_class_Equity -0.4496 0.058 -7.785 0.000 -0.563 -0.336\n", "asset_class_Fixed_Income -0.6544 0.061 -10.670 0.000 -0.775 -0.534\n", "==============================================================================\n", "Omnibus: 210.403 Durbin-Watson: 1.932\n", "Prob(Omnibus): 0.000 Jarque-Bera (JB): 363.791\n", "Skew: 0.752 Prob(JB): 1.01e-79\n", "Kurtosis: 4.541 Cond. No. 207.\n", "==============================================================================\n", "\n", "Warnings:\n", "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n" ] }, { "output_type": "stream", "name": "stderr", "text": [ "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/tsatools.py:117: FutureWarning: In a future version of pandas all arguments of concat except for the argument 'objs' will be keyword-only\n", " x = pd.concat(x[::order], 1)\n" ] } ] }, { "cell_type": "code", "source": [ "" ], "metadata": { "id": "9bOBjhpriuqZ" }, "execution_count": null, "outputs": [] }, { "cell_type": "code", "source": [ "beta = np.linalg.inv(X_ols.T.dot(X_ols)).dot(X_ols.T.dot(y))\n", "pd.Series(beta, index=X_ols.columns)" ], "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "l6tjX3JDiI7j", "outputId": "f2d631df-d1a5-4baf-e8ae-a5d36d04e017" }, "execution_count": 185, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ "const 2.216848\n", "scaled_aum -0.069728\n", "beta 0.066850\n", "asset_class_Commodities -0.627084\n", "asset_class_Currency -0.126001\n", "asset_class_Equity -0.449633\n", "asset_class_Fixed_Income -0.654390\n", "dtype: float64" ] }, "metadata": {}, "execution_count": 185 } ] } ], "metadata": { "anaconda-cloud": {}, "colab": { "collapsed_sections": [], "name": "simple_regression_1.ipynb", "provenance": [], "include_colab_link": true }, "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": "3.9.7" } }, "nbformat": 4, "nbformat_minor": 0 }