diff --git a/notebooks/chapter2.ipynb b/notebooks/chapter2.ipynb index 3c69c83..39ff628 100644 --- a/notebooks/chapter2.ipynb +++ b/notebooks/chapter2.ipynb @@ -55,7 +55,7 @@ }, { "cell_type": "code", - "execution_count": 2, + "execution_count": 7, "id": "922747da-69f3-4536-95ab-7c06c2d3ab3a", "metadata": {}, "outputs": [], @@ -66,7 +66,7 @@ }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 8, "id": "0c64cb20-fb60-4ed1-ab15-e1a08c835bd9", "metadata": {}, "outputs": [ @@ -79,7 +79,7 @@ "Name: response_att, dtype: float64" ] }, - "execution_count": 3, + "execution_count": 8, "metadata": {}, "output_type": "execute_result" } @@ -98,7 +98,7 @@ }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 9, "id": "0fe4b5ff-1a00-47d3-83b3-6d7533e1e8b3", "metadata": {}, "outputs": [ @@ -108,13 +108,13 @@ "" ] }, - "execution_count": 4, + "execution_count": 9, "metadata": {}, "output_type": "execute_result" }, { "data": { - "image/png": "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\n", + "image/png": "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", "text/plain": [ "
" ] @@ -137,7 +137,7 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": 10, "id": "91f9ee6b-6d05-4e2b-b201-0cf566c4d1b6", "metadata": {}, "outputs": [ @@ -147,13 +147,13 @@ "" ] }, - "execution_count": 5, + "execution_count": 10, "metadata": {}, "output_type": "execute_result" }, { "data": { - "image/png": "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\n", + "image/png": "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", "text/plain": [ "
" ] @@ -174,7 +174,7 @@ }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 11, "id": "6b83ccc6-1f8a-4e5b-85b9-58cb346019b7", "metadata": {}, "outputs": [ @@ -187,7 +187,7 @@ "Name: food_share_15d, dtype: float64" ] }, - "execution_count": 6, + "execution_count": 11, "metadata": {}, "output_type": "execute_result" } @@ -226,28 +226,28 @@ }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 20, "id": "83cb481a-f1a0-4620-80cf-bcfa98b358ff", "metadata": {}, "outputs": [], "source": [ - "rng = np.random.default_rng()\n", + "rng = np.random.default_rng() #[nits] ここはシード固定はしなくていい? rng = np.random.default_rng(seed=0)\n", "is_treatment = rng.choice([0, 1], p=[0.3, 0.7])" ] }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 22, "id": "36a395e5-c923-4140-bd74-6b13e89a511a", "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "0" + "1" ] }, - "execution_count": 8, + "execution_count": 22, "metadata": {}, "output_type": "execute_result" } @@ -281,7 +281,7 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 40, "id": "7be0da4e-7437-4eea-81ac-e5f0c13122ef", "metadata": {}, "outputs": [], @@ -298,7 +298,7 @@ }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 41, "id": "4cad28ee-b77a-4ddc-bb72-fcae048a0b31", "metadata": {}, "outputs": [ @@ -308,7 +308,7 @@ "1" ] }, - "execution_count": 10, + "execution_count": 41, "metadata": {}, "output_type": "execute_result" } @@ -335,7 +335,7 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 44, "id": "6200c5c4-4f2a-49d9-9552-451464f7fb2c", "metadata": {}, "outputs": [ @@ -355,13 +355,13 @@ "" ] }, - "execution_count": 11, + "execution_count": 44, "metadata": {}, "output_type": "execute_result" }, { "data": { - "image/png": "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\n", + "image/png": "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", "text/plain": [ "
" ] @@ -421,7 +421,7 @@ }, { "cell_type": "code", - "execution_count": 12, + "execution_count": 45, "id": "21933ecb-0289-4855-a820-dee853a4b7f4", "metadata": {}, "outputs": [ @@ -440,10 +440,10 @@ " Method: Least Squares F-statistic: 7.890 \n", "\n", "\n", - " Date: Wed, 14 Feb 2024 Prob (F-statistic): 0.00497 \n", + " Date: Mon, 26 Feb 2024 Prob (F-statistic): 0.00497 \n", "\n", "\n", - " Time: 01:18:51 Log-Likelihood: -12692. \n", + " Time: 14:38:13 Log-Likelihood: -12692. \n", "\n", "\n", " No. Observations: 50000 AIC: 2.539e+04\n", @@ -491,8 +491,8 @@ "\\textbf{Dep. Variable:} & response\\_att & \\textbf{ R-squared: } & 0.000 \\\\\n", "\\textbf{Model:} & OLS & \\textbf{ Adj. R-squared: } & 0.000 \\\\\n", "\\textbf{Method:} & Least Squares & \\textbf{ F-statistic: } & 7.890 \\\\\n", - "\\textbf{Date:} & Wed, 14 Feb 2024 & \\textbf{ Prob (F-statistic):} & 0.00497 \\\\\n", - "\\textbf{Time:} & 01:18:51 & \\textbf{ Log-Likelihood: } & -12692. \\\\\n", + "\\textbf{Date:} & Mon, 26 Feb 2024 & \\textbf{ Prob (F-statistic):} & 0.00497 \\\\\n", + "\\textbf{Time:} & 14:38:13 & \\textbf{ Log-Likelihood: } & -12692. \\\\\n", "\\textbf{No. Observations:} & 50000 & \\textbf{ AIC: } & 2.539e+04 \\\\\n", "\\textbf{Df Residuals:} & 49998 & \\textbf{ BIC: } & 2.541e+04 \\\\\n", "\\textbf{Df Model:} & 1 & \\textbf{ } & \\\\\n", @@ -527,8 +527,8 @@ "Dep. Variable: response_att R-squared: 0.000\n", "Model: OLS Adj. R-squared: 0.000\n", "Method: Least Squares F-statistic: 7.890\n", - "Date: Wed, 14 Feb 2024 Prob (F-statistic): 0.00497\n", - "Time: 01:18:51 Log-Likelihood: -12692.\n", + "Date: Mon, 26 Feb 2024 Prob (F-statistic): 0.00497\n", + "Time: 14:38:13 Log-Likelihood: -12692.\n", "No. Observations: 50000 AIC: 2.539e+04\n", "Df Residuals: 49998 BIC: 2.541e+04\n", "Df Model: 1 \n", @@ -550,7 +550,7 @@ "\"\"\"" ] }, - "execution_count": 12, + "execution_count": 45, "metadata": {}, "output_type": "execute_result" } @@ -569,7 +569,7 @@ }, { "cell_type": "code", - "execution_count": 13, + "execution_count": 46, "id": "de7df562-4adc-4985-a10b-5cc26f537676", "metadata": {}, "outputs": [ @@ -655,7 +655,7 @@ "4 1 0 0.5105 20.0 1" ] }, - "execution_count": 13, + "execution_count": 46, "metadata": {}, "output_type": "execute_result" } @@ -666,7 +666,7 @@ }, { "cell_type": "code", - "execution_count": 14, + "execution_count": 47, "id": "d2cf0a62-bb55-4d86-b5ad-b5f7ff24850f", "metadata": {}, "outputs": [ @@ -726,7 +726,7 @@ "1 0.348872 43.803353 0.629673" ] }, - "execution_count": 14, + "execution_count": 47, "metadata": {}, "output_type": "execute_result" } @@ -920,7 +920,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.8.5" + "version": "3.9.6" } }, "nbformat": 4, diff --git a/notebooks/chapter3_abtest_detail.ipynb b/notebooks/chapter3_abtest_detail.ipynb index 64d567a..46c5b42 100644 --- a/notebooks/chapter3_abtest_detail.ipynb +++ b/notebooks/chapter3_abtest_detail.ipynb @@ -20,32 +20,43 @@ "name": "stdout", "output_type": "stream", "text": [ - "Requirement already satisfied: linearmodels in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (4.31)\n", - "Requirement already satisfied: Cython>=0.29.21 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from linearmodels) (3.0.2)\n", - "Requirement already satisfied: property-cached>=1.6.3 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from linearmodels) (1.6.4)\n", - "Requirement already satisfied: setuptools-scm[toml]>=7.0.0 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from linearmodels) (7.1.0)\n", - "Requirement already satisfied: numpy>=1.16 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from linearmodels) (1.24.4)\n", - "Requirement already satisfied: formulaic>=0.3.2 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from linearmodels) (0.6.4)\n", - "Requirement already satisfied: pandas>=0.24 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from linearmodels) (2.0.3)\n", - "Requirement already satisfied: pyhdfe>=0.1 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from linearmodels) (0.2.0)\n", - "Requirement already satisfied: scipy>=1.2 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from linearmodels) (1.10.1)\n", - "Requirement already satisfied: mypy-extensions>=0.4 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from linearmodels) (1.0.0)\n", - "Requirement already satisfied: statsmodels>=0.11 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from linearmodels) (0.14.1)\n", - "Requirement already satisfied: wrapt>=1.0 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from formulaic>=0.3.2->linearmodels) (1.15.0)\n", - "Requirement already satisfied: interface-meta>=1.2.0 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from formulaic>=0.3.2->linearmodels) (1.3.0)\n", - "Requirement already satisfied: typing-extensions>=4.2.0 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from formulaic>=0.3.2->linearmodels) (4.8.0)\n", - "Requirement already satisfied: astor>=0.8 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from formulaic>=0.3.2->linearmodels) (0.8.1)\n", - "Requirement already satisfied: graphlib-backport>=1.0.0 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from formulaic>=0.3.2->linearmodels) (1.0.3)\n", - "Requirement already satisfied: pytz>=2020.1 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from pandas>=0.24->linearmodels) (2023.3.post1)\n", - "Requirement already satisfied: tzdata>=2022.1 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from pandas>=0.24->linearmodels) (2023.4)\n", - "Requirement already satisfied: python-dateutil>=2.8.2 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from pandas>=0.24->linearmodels) (2.8.2)\n", - "Requirement already satisfied: six>=1.5 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from python-dateutil>=2.8.2->pandas>=0.24->linearmodels) (1.16.0)\n", - "Requirement already satisfied: packaging>=20.0 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from setuptools-scm[toml]>=7.0.0->linearmodels) (23.1)\n", - "Requirement already satisfied: tomli>=1.0.0 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from setuptools-scm[toml]>=7.0.0->linearmodels) (2.0.1)\n", - "Requirement already satisfied: setuptools in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from setuptools-scm[toml]>=7.0.0->linearmodels) (52.0.0)\n", - "Requirement already satisfied: patsy>=0.5.4 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from statsmodels>=0.11->linearmodels) (0.5.6)\n", - "\u001b[33mWARNING: You are using pip version 21.0.1; however, version 24.0 is available.\n", - "You should consider upgrading via the '/Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/bin/python -m pip install --upgrade pip' command.\u001b[0m\n" + "Collecting linearmodels\n", + " Downloading linearmodels-5.4-cp39-cp39-macosx_11_0_arm64.whl (1.5 MB)\n", + "\u001b[2K \u001b[38;2;114;156;31m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.5/1.5 MB\u001b[0m \u001b[31m6.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m4 MB/s\u001b[0m eta \u001b[36m0:00:01\u001b[0m\n", + "\u001b[?25hRequirement already satisfied: scipy>=1.5.0 in /Users/s01057/mylocal/lib/python3.9/site-packages (from linearmodels) (1.11.3)\n", + "Requirement already satisfied: statsmodels>=0.12.0 in /Users/s01057/mylocal/lib/python3.9/site-packages (from linearmodels) (0.14.1)\n", + "Collecting formulaic>=0.6.5\n", + " Downloading formulaic-1.0.1-py3-none-any.whl (94 kB)\n", + "\u001b[2K \u001b[38;2;114;156;31m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m94.2/94.2 kB\u001b[0m \u001b[31m9.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[?25hCollecting mypy-extensions>=0.4\n", + " Downloading mypy_extensions-1.0.0-py3-none-any.whl (4.7 kB)\n", + "Requirement already satisfied: pandas>=1.3.0 in /Users/s01057/mylocal/lib/python3.9/site-packages (from linearmodels) (2.1.4)\n", + "Requirement already satisfied: numpy>=1.22.0 in /Users/s01057/mylocal/lib/python3.9/site-packages (from linearmodels) (1.26.1)\n", + "Collecting pyhdfe>=0.1\n", + " Downloading pyhdfe-0.2.0-py3-none-any.whl (19 kB)\n", + "Collecting Cython>=0.29.37\n", + " Downloading Cython-3.0.8-py2.py3-none-any.whl (1.2 MB)\n", + "\u001b[2K \u001b[38;2;114;156;31m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.2/1.2 MB\u001b[0m \u001b[31m31.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[?25hCollecting setuptools-scm[toml]<9.0.0,>=8.0.0\n", + " Downloading setuptools_scm-8.0.4-py3-none-any.whl (42 kB)\n", + "\u001b[2K \u001b[38;2;114;156;31m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m42.1/42.1 kB\u001b[0m \u001b[31m4.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[?25hRequirement already satisfied: typing-extensions>=4.2.0 in /Users/s01057/mylocal/lib/python3.9/site-packages (from formulaic>=0.6.5->linearmodels) (4.8.0)\n", + "Collecting interface-meta>=1.2.0\n", + " Downloading interface_meta-1.3.0-py3-none-any.whl (14 kB)\n", + "Collecting wrapt>=1.0\n", + " Downloading wrapt-1.16.0-cp39-cp39-macosx_11_0_arm64.whl (38 kB)\n", + "Requirement already satisfied: python-dateutil>=2.8.2 in /Users/s01057/mylocal/lib/python3.9/site-packages (from pandas>=1.3.0->linearmodels) (2.8.2)\n", + "Requirement already satisfied: tzdata>=2022.1 in /Users/s01057/mylocal/lib/python3.9/site-packages (from pandas>=1.3.0->linearmodels) (2023.3)\n", + "Requirement already satisfied: pytz>=2020.1 in /Users/s01057/mylocal/lib/python3.9/site-packages (from pandas>=1.3.0->linearmodels) (2023.3.post1)\n", + "Requirement already satisfied: packaging>=20 in /Users/s01057/mylocal/lib/python3.9/site-packages (from setuptools-scm[toml]<9.0.0,>=8.0.0->linearmodels) (23.2)\n", + "Requirement already satisfied: tomli>=1 in /Users/s01057/mylocal/lib/python3.9/site-packages (from setuptools-scm[toml]<9.0.0,>=8.0.0->linearmodels) (2.0.1)\n", + "Requirement already satisfied: setuptools in /Users/s01057/mylocal/lib/python3.9/site-packages (from setuptools-scm[toml]<9.0.0,>=8.0.0->linearmodels) (58.0.4)\n", + "Requirement already satisfied: patsy>=0.5.4 in /Users/s01057/mylocal/lib/python3.9/site-packages (from statsmodels>=0.12.0->linearmodels) (0.5.6)\n", + "Requirement already satisfied: six in /Users/s01057/mylocal/lib/python3.9/site-packages (from patsy>=0.5.4->statsmodels>=0.12.0->linearmodels) (1.16.0)\n", + "Installing collected packages: wrapt, setuptools-scm, mypy-extensions, interface-meta, Cython, pyhdfe, formulaic, linearmodels\n", + "Successfully installed Cython-3.0.8 formulaic-1.0.1 interface-meta-1.3.0 linearmodels-5.4 mypy-extensions-1.0.0 pyhdfe-0.2.0 setuptools-scm-8.0.4 wrapt-1.16.0\n", + "\u001b[33mWARNING: There was an error checking the latest version of pip.\u001b[0m\u001b[33m\n", + "\u001b[0m" ] } ], @@ -118,7 +129,7 @@ }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 4, "id": "cbe4fade-1a09-4360-9b97-9fea0ce791c2", "metadata": {}, "outputs": [ @@ -137,10 +148,10 @@ " Method: Least Squares F-statistic: 5.594 \n", "\n", "\n", - " Date: Wed, 14 Feb 2024 Prob (F-statistic): 0.0180 \n", + " Date: Mon, 26 Feb 2024 Prob (F-statistic): 0.0180 \n", "\n", "\n", - " Time: 01:17:18 Log-Likelihood: -7718.2 \n", + " Time: 14:40:23 Log-Likelihood: -7718.2 \n", "\n", "\n", " No. Observations: 10640 AIC: 1.544e+04\n", @@ -188,8 +199,8 @@ "\\textbf{Dep. Variable:} & is\\_click & \\textbf{ R-squared: } & 0.001 \\\\\n", "\\textbf{Model:} & OLS & \\textbf{ Adj. R-squared: } & 0.000 \\\\\n", "\\textbf{Method:} & Least Squares & \\textbf{ F-statistic: } & 5.594 \\\\\n", - "\\textbf{Date:} & Wed, 14 Feb 2024 & \\textbf{ Prob (F-statistic):} & 0.0180 \\\\\n", - "\\textbf{Time:} & 01:17:18 & \\textbf{ Log-Likelihood: } & -7718.2 \\\\\n", + "\\textbf{Date:} & Mon, 26 Feb 2024 & \\textbf{ Prob (F-statistic):} & 0.0180 \\\\\n", + "\\textbf{Time:} & 14:40:23 & \\textbf{ Log-Likelihood: } & -7718.2 \\\\\n", "\\textbf{No. Observations:} & 10640 & \\textbf{ AIC: } & 1.544e+04 \\\\\n", "\\textbf{Df Residuals:} & 10638 & \\textbf{ BIC: } & 1.546e+04 \\\\\n", "\\textbf{Df Model:} & 1 & \\textbf{ } & \\\\\n", @@ -224,8 +235,8 @@ "Dep. Variable: is_click R-squared: 0.001\n", "Model: OLS Adj. R-squared: 0.000\n", "Method: Least Squares F-statistic: 5.594\n", - "Date: Wed, 14 Feb 2024 Prob (F-statistic): 0.0180\n", - "Time: 01:17:18 Log-Likelihood: -7718.2\n", + "Date: Mon, 26 Feb 2024 Prob (F-statistic): 0.0180\n", + "Time: 14:40:23 Log-Likelihood: -7718.2\n", "No. Observations: 10640 AIC: 1.544e+04\n", "Df Residuals: 10638 BIC: 1.546e+04\n", "Df Model: 1 \n", @@ -247,7 +258,7 @@ "\"\"\"" ] }, - "execution_count": 3, + "execution_count": 4, "metadata": {}, "output_type": "execute_result" } @@ -262,7 +273,7 @@ }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 5, "id": "48d0e6e8-de6f-452b-88d1-ab4d2decdad5", "metadata": {}, "outputs": [ @@ -381,7 +392,7 @@ "[10640 rows x 3 columns]" ] }, - "execution_count": 4, + "execution_count": 5, "metadata": {}, "output_type": "execute_result" } @@ -409,7 +420,7 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": 6, "id": "478346d1-c7f5-4582-a5b7-2674eb3b06ab", "metadata": {}, "outputs": [ @@ -444,7 +455,7 @@ "" ] }, - "execution_count": 5, + "execution_count": 6, "metadata": {}, "output_type": "execute_result" } @@ -459,7 +470,7 @@ }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 7, "id": "6f2e4b2d-c316-4e1e-9de5-b63e8335f8aa", "metadata": {}, "outputs": [ @@ -578,7 +589,7 @@ "[10000 rows x 3 columns]" ] }, - "execution_count": 6, + "execution_count": 7, "metadata": {}, "output_type": "execute_result" } @@ -637,7 +648,7 @@ }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 14, "id": "dff7f3ff-e84d-429e-8e4d-714c4cf7ca42", "metadata": {}, "outputs": [ @@ -645,7 +656,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "100%|█████████████████████████████████████████████| 300/300 [00:10<00:00, 28.60it/s]\n" + "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 300/300 [00:05<00:00, 50.39it/s]\n" ] } ], @@ -698,13 +709,13 @@ }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 15, "id": "6abfb07b-bc98-4c57-9a03-44c4b2557355", "metadata": {}, "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "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", "text/plain": [ "
" ] @@ -738,17 +749,17 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 16, "id": "b702f169-9471-4d72-aab7-032b60edae72", "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "KstestResult(statistic=0.04504551195892681, pvalue=0.5611215508935568, statistic_location=0.6883788452922601, statistic_sign=-1)" + "KstestResult(statistic=0.04504551195894413, pvalue=0.5611215508930631, statistic_location=0.6883788452922774, statistic_sign=-1)" ] }, - "execution_count": 9, + "execution_count": 16, "metadata": {}, "output_type": "execute_result" } @@ -816,7 +827,7 @@ }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 17, "id": "981020b9-47ab-4adf-8dfe-d481650eeef3", "metadata": {}, "outputs": [ @@ -824,7 +835,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "100%|█████████████████████████████████████████████| 300/300 [00:11<00:00, 26.00it/s]\n" + "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 300/300 [00:06<00:00, 49.24it/s]\n" ] } ], @@ -856,7 +867,7 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 18, "id": "456e6894-67b4-475d-9564-e7b118d612e1", "metadata": {}, "outputs": [ @@ -987,7 +998,7 @@ "[10640 rows x 4 columns]" ] }, - "execution_count": 11, + "execution_count": 18, "metadata": {}, "output_type": "execute_result" } @@ -998,13 +1009,13 @@ }, { "cell_type": "code", - "execution_count": 12, + "execution_count": 19, "id": "6158565e-0a5d-4a6e-8de0-c6cf6088b38c", "metadata": {}, "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "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", "text/plain": [ "
" ] @@ -1024,17 +1035,17 @@ }, { "cell_type": "code", - "execution_count": 13, + "execution_count": 20, "id": "4738caac-3692-4d5e-a203-1bcbd36c96bd", "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "KstestResult(statistic=0.6134127868406709, pvalue=4.70019809957193e-109, statistic_location=0.056587213159329075, statistic_sign=1)" + "KstestResult(statistic=0.6134127868406839, pvalue=4.700198099515917e-109, statistic_location=0.05658721315931616, statistic_sign=1)" ] }, - "execution_count": 13, + "execution_count": 20, "metadata": {}, "output_type": "execute_result" } @@ -1066,7 +1077,7 @@ }, { "cell_type": "code", - "execution_count": 14, + "execution_count": 22, "id": "4f9b7c6a-64b6-44e8-93c2-e03ab07f7a9e", "metadata": {}, "outputs": [ @@ -1101,7 +1112,7 @@ "" ] }, - "execution_count": 14, + "execution_count": 22, "metadata": {}, "output_type": "execute_result" } @@ -1121,7 +1132,7 @@ }, { "cell_type": "code", - "execution_count": 15, + "execution_count": 23, "id": "26bb5ad6-d815-4755-abfc-3c8f31a26250", "metadata": {}, "outputs": [ @@ -1240,7 +1251,7 @@ "[10640 rows x 3 columns]" ] }, - "execution_count": 15, + "execution_count": 23, "metadata": {}, "output_type": "execute_result" } @@ -1289,7 +1300,7 @@ }, { "cell_type": "code", - "execution_count": 16, + "execution_count": 24, "id": "e95ae530-aa40-47c1-a968-0099fc1a5a1e", "metadata": {}, "outputs": [ @@ -1297,7 +1308,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "100%|█████████████████████████████████████████████| 300/300 [00:12<00:00, 24.16it/s]\n" + "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 300/300 [00:06<00:00, 44.96it/s]\n" ] } ], @@ -1333,13 +1344,13 @@ }, { "cell_type": "code", - "execution_count": 17, + "execution_count": 25, "id": "0a06f42f-a4ca-432e-b365-a005a520c754", "metadata": {}, "outputs": [ { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbIAAAGJCAYAAADvzrZyAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/OQEPoAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAtuUlEQVR4nO3deVRV9f7/8dcB5ODAICpT4mw5ot8olRxzQvRa3viWpSvRZWmFVvK7N6MJtQGzvmrXqzRo2m1plq3MTNPUEq8mZippaaQ40VfBoQRFRYTP74++nOtRUA8cxK3Px1p7Lc5nf/Znv88HPC/3cM6xGWOMAACwKI+qLgAAgIogyAAAlkaQAQAsjSADAFgaQQYAsDSCDABgaQQZAMDSCDIAgKURZAAASyPIUCkmTJggm83m1NaoUSMNHz680ve9f/9+2Ww2zZs3z9E2fPhw1apVq9L3XcJms2nChAnXbH/lsXnzZt11112qWbOmbDab0tPTq7qkS/To0UM9evSo6jJwnSPIcF1bvnz5dRsI13NtV1JYWKj7779fv//+u6ZNm6YPP/xQDRs2rOqygHLxquoCcPPIyMiQh4dr/3davny5Zs6c6VJgNGzYUGfOnFG1atVcrNA1l6vtzJkz8vK6fv95ZWZm6sCBA3rvvff0yCOPVHU5QIVcv//ScMOx2+2VOv758+dVXFwsb29v+fj4VOq+rqSq938lR44ckSQFBARUbSGAG3BqERW2fv163XnnnfLx8VHTpk31zjvvlNrv4mtkhYWFmjhxopo3by4fHx/VqVNHXbp00apVqyT9eV1r5syZkv685lSySP+5Dvbmm29q+vTpatq0qex2u3bu3FnqNbISe/fuVXR0tGrWrKmwsDBNmjRJF34BxNq1a2Wz2bR27Vqn7S4e83K1lbRdfKS2bds2xcTEyM/PT7Vq1VKvXr2Ulpbm1GfevHmy2WzasGGDEhISVK9ePdWsWVN//etfdfTo0dJ/ARf55ptv1LVrV9WsWVMBAQG69957tWvXLsf64cOHq3v37pKk+++/Xzab7bLXoUpqWrdunUaPHq06derIz89Pw4YN0x9//OHo95e//EVNmjQpdYyoqCjdcccdjsdz585Vz549FRQUJLvdrlatWiklJeWKz62klv379zu1l/V727Rpk/r16yd/f3/VqFFD3bt314YNG664H1gLR2SokB07dqhv376qV6+eJkyYoPPnzyspKUnBwcFX3HbChAlKTk7WI488og4dOigvL08//PCDtm7dqj59+mj06NE6dOiQVq1apQ8//LDUMebOnauzZ89q1KhRstvtCgwMVHFxcal9i4qK1K9fP3Xq1ElTpkzRihUrlJSUpPPnz2vSpEkuPe+rqe1CP//8s7p27So/Pz8988wzqlatmt555x316NFDqamp6tixo1P/sWPHqnbt2kpKStL+/fs1ffp0jRkzRh9//PFl97N69WrFxMSoSZMmmjBhgs6cOaMZM2aoc+fO2rp1qxo1aqTRo0frlltu0WuvvaYnn3xSd95551X9vsaMGaOAgABNmDBBGRkZSklJ0YEDBxwhMnjwYA0bNkybN2/WnXfe6djuwIEDSktL0xtvvOFoS0lJUevWrXXPPffIy8tLS5cu1RNPPKHi4mLFx8dfsZar8c033ygmJkaRkZFKSkqSh4eHI0D//e9/q0OHDm7ZD64DBqiAQYMGGR8fH3PgwAFH286dO42np6e5+M+rYcOGJi4uzvG4Xbt2ZsCAAZcdPz4+/pJxjDFm3759RpLx8/MzR44cKXXd3LlzHW1xcXFGkhk7dqyjrbi42AwYMMB4e3ubo0ePGmOM+fbbb40k8+23315xzLJqM8YYSSYpKcnxeNCgQcbb29tkZmY62g4dOmR8fX1Nt27dHG1z5841kkzv3r1NcXGxo33cuHHG09PTnDhxotT9lWjfvr0JCgoyx48fd7T9+OOPxsPDwwwbNszRVvI8Fy1adNnxLqwpMjLSnDt3ztE+ZcoUI8ksWbLEGGNMbm6usdvt5v/9v//ntP2UKVOMzWZz+hs5ffr0JfuJjo42TZo0cWrr3r276d69+yW17Nu3z6nfxb+34uJi07x5cxMdHe00j6dPnzaNGzc2ffr0ueLzhnVwahHlVlRUpJUrV2rQoEFq0KCBo71ly5aKjo6+4vYBAQH6+eeftXv37nLXEBsbq3r16l11/zFjxjh+ttlsGjNmjM6dO6fVq1eXu4YrKSoq0tdff61BgwY5nXoLDQ3VkCFDtH79euXl5TltM2rUKKdTlV27dlVRUZEOHDhQ5n4OHz6s9PR0DR8+XIGBgY72iIgI9enTR8uXL6/Q8xg1apTTDTSPP/64vLy8HOP6+fkpJiZGn3zyidPp2o8//lidOnVy+hupXr264+fc3FwdO3ZM3bt31969e5Wbm1uhOiUpPT1du3fv1pAhQ3T8+HEdO3ZMx44dU35+vnr16qV169aVeeQO6yHIUG5Hjx7VmTNn1Lx580vW3XbbbVfcftKkSTpx4oRuvfVWtW3bVn//+9+1fft2l2po3LjxVff18PC45BrOrbfeKkmXXHNxp6NHj+r06dOlzknLli1VXFysrKwsp/YLX/QlqXbt2pLkdE3qYiUhV9Z+Sl7Iy+vi33OtWrUUGhrqNHeDBw9WVlaWNm7cKOnPuyO3bNmiwYMHO227YcMG9e7d23Edr169enruueckyS1BVvKfo7i4ONWrV89pmT17tgoKCtyyH1wfuEaGKtOtWzdlZmZqyZIl+vrrrzV79mxNmzZNb7/99lXfEn7h/+zd4eI3cZcoKipy636uxNPTs9T2C490rkcDBw5UjRo19Mknn+iuu+7SJ598Ig8PD91///2OPpmZmerVq5datGihqVOnKjw8XN7e3lq+fLmmTZt22SOlq/39lIzxxhtvqH379qVucy3fII/KRZCh3OrVq6fq1auXemowIyPjqsYIDAzUiBEjNGLECJ06dUrdunXThAkTHEFW1gtXeRQXF2vv3r2OozBJ+vXXXyX9eUel9J8jnxMnTjhtW9opvautrV69eqpRo0apc/LLL7/Iw8ND4eHhVzXW5ZS8obms/dStW1c1a9Ys9/i7d+/W3Xff7Xh86tQpHT58WP3793e01axZU3/5y1+0aNEiTZ06VR9//LG6du2qsLAwR5+lS5eqoKBAX3zxhdOR57fffnvFGq7299O0aVNJf57u7N2799U/SVgSpxZRbp6enoqOjtbnn3+ugwcPOtp37dqllStXXnH748ePOz2uVauWmjVrpoKCAkdbyQvvxS9c5fXPf/7T8bMxRv/85z9VrVo19erVS9KfYeDp6al169Y5bTdr1qxLxrra2jw9PdW3b18tWbLE6TRcTk6OFixYoC5dusjPz6+cz+g/QkND1b59e33wwQdONf3000/6+uuvnQKnPN59910VFhY6HqekpOj8+fOKiYlx6jd48GAdOnRIs2fP1o8//njJacWSo80Ljy5zc3M1d+7cK9ZQElAX/n6Kior07rvvOvWLjIxU06ZN9eabb+rUqVOXjHO1b2WANXBEhgqZOHGiVqxYoa5du+qJJ57Q+fPnNWPGDLVu3fqK17tatWqlHj16KDIyUoGBgfrhhx/06aefOt2QERkZKUl68sknFR0dLU9PTz344IPlqtXHx0crVqxQXFycOnbsqK+++krLli3Tc88957hhxN/fX/fff79mzJghm82mpk2b6ssvv3S8gfhCrtT2yiuvaNWqVerSpYueeOIJeXl56Z133lFBQYGmTJlSrudTmjfeeEMxMTGKiorSyJEjHbff+/v7V/jjtM6dO6devXrpgQceUEZGhmbNmqUuXbronnvucerXv39/+fr66m9/+5s8PT0VGxvrtL5v377y9vbWwIEDNXr0aJ06dUrvvfeegoKCdPjw4cvW0Lp1a3Xq1EmJiYn6/fffFRgYqIULF+r8+fNO/Tw8PDR79mzFxMSodevWGjFihG655Rb97//+r7799lv5+flp6dKlFZoPXEeq9qZJ3AhSU1NNZGSk8fb2Nk2aNDFvv/22SUpKuuLt96+88orp0KGDCQgIMNWrVzctWrQwr776qtMt3ufPnzdjx4419erVMzabzTFmye3wb7zxxiX1lHX7fc2aNU1mZqbp27evqVGjhgkODjZJSUmmqKjIafujR4+a2NhYU6NGDVO7dm0zevRo89NPP10yZlm1GXPp7ffGGLN161YTHR1tatWqZWrUqGHuvvtu89133zn1Kbm9fPPmzU7tZb0toDSrV682nTt3NtWrVzd+fn5m4MCBZufOnaWO58rt96mpqWbUqFGmdu3aplatWmbo0KFOt/lfaOjQoY63EZTmiy++MBEREcbHx8c0atTIvP766+b999+/5Nb6i2+/N8aYzMxM07t3b2O3201wcLB57rnnzKpVq0qdn23btpn77rvP1KlTx9jtdtOwYUPzwAMPmDVr1lzxecM6bMZc51ePAVSpefPmacSIEdq8ebPTp3MA1wuukQEALI0gAwBYGkEGALA0rpEBACyNIzIAgKURZAAAS7vu3hBdXFysQ4cOydfX160fTwQAsBZjjE6ePKmwsDB5eJR93HXdBdmhQ4fc8rlzAIAbQ1ZWlurXr1/m+usuyHx9fSX9Wbg7Pn8OAGBNeXl5Cg8Pd+RCWa67ICs5nejn50eQAQCueJmJmz0AAJZGkAEALI0gAwBYGkEGALA0ggwAYGkEGQDA0ggyAIClEWQAAEsjyAAAlkaQAQAsjSADAFjadfdZi+7U6NllVV2Cw/7JA6q6BAC4IXFEBgCwNIIMAGBpBBkAwNIqFGSTJ0+WzWbT008/7Wg7e/as4uPjVadOHdWqVUuxsbHKycmpaJ0AAJSq3EG2efNmvfPOO4qIiHBqHzdunJYuXapFixYpNTVVhw4d0n333VfhQgEAKE25guzUqVMaOnSo3nvvPdWuXdvRnpubqzlz5mjq1Knq2bOnIiMjNXfuXH333XdKS0tzW9EAAJQoV5DFx8drwIAB6t27t1P7li1bVFhY6NTeokULNWjQQBs3bix1rIKCAuXl5TktAABcLZffR7Zw4UJt3bpVmzdvvmRddna2vL29FRAQ4NQeHBys7OzsUsdLTk7WxIkTXS0DAABJLh6RZWVl6amnntL8+fPl4+PjlgISExOVm5vrWLKystwyLgDg5uBSkG3ZskVHjhzR7bffLi8vL3l5eSk1NVX/+Mc/5OXlpeDgYJ07d04nTpxw2i4nJ0chISGljmm32+Xn5+e0AABwtVw6tdirVy/t2LHDqW3EiBFq0aKFxo8fr/DwcFWrVk1r1qxRbGysJCkjI0MHDx5UVFSU+6oGAOD/uBRkvr6+atOmjVNbzZo1VadOHUf7yJEjlZCQoMDAQPn5+Wns2LGKiopSp06d3Fc1AAD/x+0fGjxt2jR5eHgoNjZWBQUFio6O1qxZs9y9GwAAJEk2Y4yp6iIulJeXJ39/f+Xm5lb4ehmffg8A1nW1ecBnLQIALI0gAwBYGkEGALA0ggwAYGkEGQDA0ggyAIClEWQAAEsjyAAAlkaQAQAsjSADAFgaQQYAsDSCDABgaQQZAMDSCDIAgKURZAAASyPIAACWRpABACyNIAMAWBpBBgCwNIIMAGBpBBkAwNK8qroAAHBVo2eXVXUJDvsnD6jqEm56HJEBACyNIAMAWJpLQZaSkqKIiAj5+fnJz89PUVFR+uqrrxzre/ToIZvN5rQ89thjbi8aAIASLl0jq1+/viZPnqzmzZvLGKMPPvhA9957r7Zt26bWrVtLkh599FFNmjTJsU2NGjXcWzEAABdwKcgGDhzo9PjVV19VSkqK0tLSHEFWo0YNhYSEuK9CAAAuo9zXyIqKirRw4ULl5+crKirK0T5//nzVrVtXbdq0UWJiok6fPn3ZcQoKCpSXl+e0AABwtVy+/X7Hjh2KiorS2bNnVatWLS1evFitWrWSJA0ZMkQNGzZUWFiYtm/frvHjxysjI0OfffZZmeMlJydr4sSJ5X8GAICbmstBdttttyk9PV25ubn69NNPFRcXp9TUVLVq1UqjRo1y9Gvbtq1CQ0PVq1cvZWZmqmnTpqWOl5iYqISEBMfjvLw8hYeHl+OpAABuRi4Hmbe3t5o1ayZJioyM1ObNm/XWW2/pnXfeuaRvx44dJUl79uwpM8jsdrvsdrurZQAAIMkN7yMrLi5WQUFBqevS09MlSaGhoRXdDQAApXLpiCwxMVExMTFq0KCBTp48qQULFmjt2rVauXKlMjMztWDBAvXv31916tTR9u3bNW7cOHXr1k0RERGVVT8A4CbnUpAdOXJEw4YN0+HDh+Xv76+IiAitXLlSffr0UVZWllavXq3p06crPz9f4eHhio2N1QsvvFBZtQMA4FqQzZkzp8x14eHhSk1NrXBBAAC4gs9aBABYGkEGALA0ggwAYGkEGQDA0ggyAIClEWQAAEsjyAAAlkaQAQAszeUPDQbcrdGzy6q6BIf9kwdUdQkAXMQRGQDA0ggyAIClEWQAAEsjyAAAlkaQAQAsjSADAFgaQQYAsDSCDABgaQQZAMDSCDIAgKURZAAASyPIAACWRpABACyNIAMAWBpBBgCwNIIMAGBpLgVZSkqKIiIi5OfnJz8/P0VFRemrr75yrD979qzi4+NVp04d1apVS7GxscrJyXF70QAAlHApyOrXr6/Jkydry5Yt+uGHH9SzZ0/de++9+vnnnyVJ48aN09KlS7Vo0SKlpqbq0KFDuu+++yqlcAAAJMnLlc4DBw50evzqq68qJSVFaWlpql+/vubMmaMFCxaoZ8+ekqS5c+eqZcuWSktLU6dOndxXNQAA/6fc18iKioq0cOFC5efnKyoqSlu2bFFhYaF69+7t6NOiRQs1aNBAGzduLHOcgoIC5eXlOS0AAFwtl47IJGnHjh2KiorS2bNnVatWLS1evFitWrVSenq6vL29FRAQ4NQ/ODhY2dnZZY6XnJysiRMnulw4AFwPGj27rKpLcNg/eUBVl1AlXD4iu+2225Senq5Nmzbp8ccfV1xcnHbu3FnuAhITE5Wbm+tYsrKyyj0WAODm4/IRmbe3t5o1ayZJioyM1ObNm/XWW29p8ODBOnfunE6cOOF0VJaTk6OQkJAyx7Pb7bLb7a5XDgCA3PA+suLiYhUUFCgyMlLVqlXTmjVrHOsyMjJ08OBBRUVFVXQ3AACUyqUjssTERMXExKhBgwY6efKkFixYoLVr12rlypXy9/fXyJEjlZCQoMDAQPn5+Wns2LGKiorijkUAQKVxKciOHDmiYcOG6fDhw/L391dERIRWrlypPn36SJKmTZsmDw8PxcbGqqCgQNHR0Zo1a1alFA4AgORikM2ZM+ey6318fDRz5kzNnDmzQkUBAHC1XL7ZA8DN53q6xRy4GB8aDACwNIIMAGBpBBkAwNIIMgCApRFkAABLI8gAAJbG7ffXyPV0+/LN+gnZwI3uZn2d4YgMAGBpBBkAwNIIMgCApRFkAABLI8gAAJZGkAEALI0gAwBYGkEGALA0ggwAYGkEGQDA0ggyAIClEWQAAEsjyAAAlkaQAQAsja9xuQldT1/1gLLxewKuDkdkAABLI8gAAJbmUpAlJyfrzjvvlK+vr4KCgjRo0CBlZGQ49enRo4dsNpvT8thjj7m1aAAASrgUZKmpqYqPj1daWppWrVqlwsJC9e3bV/n5+U79Hn30UR0+fNixTJkyxa1FAwBQwqWbPVasWOH0eN68eQoKCtKWLVvUrVs3R3uNGjUUEhLingoBALiMCl0jy83NlSQFBgY6tc+fP19169ZVmzZtlJiYqNOnT5c5RkFBgfLy8pwWAACuVrlvvy8uLtbTTz+tzp07q02bNo72IUOGqGHDhgoLC9P27ds1fvx4ZWRk6LPPPit1nOTkZE2cOLG8ZQAAbnLlDrL4+Hj99NNPWr9+vVP7qFGjHD+3bdtWoaGh6tWrlzIzM9W0adNLxklMTFRCQoLjcV5ensLDw8tbFgDgJlOuIBszZoy+/PJLrVu3TvXr179s344dO0qS9uzZU2qQ2e122e328pQBAIBrQWaM0dixY7V48WKtXbtWjRs3vuI26enpkqTQ0NByFQgAwOW4FGTx8fFasGCBlixZIl9fX2VnZ0uS/P39Vb16dWVmZmrBggXq37+/6tSpo+3bt2vcuHHq1q2bIiIiKuUJAABubi4FWUpKiqQ/3/R8oblz52r48OHy9vbW6tWrNX36dOXn5ys8PFyxsbF64YUX3FYwAAAXcvnU4uWEh4crNTW1QgUBAOAKPmsRAGBpBBkAwNIIMgCApRFkAABLI8gAAJZGkAEALI0gAwBYWrk/NBi4ETV6dllVlwDARRyRAQAsjSADAFgaQQYAsDSCDABgaQQZAMDSCDIAgKURZAAASyPIAACWRpABACyNIAMAWBpBBgCwNIIMAGBpBBkAwNIIMgCApRFkAABLI8gAAJZGkAEALM2lIEtOTtadd94pX19fBQUFadCgQcrIyHDqc/bsWcXHx6tOnTqqVauWYmNjlZOT49aiAQAo4VKQpaamKj4+XmlpaVq1apUKCwvVt29f5efnO/qMGzdOS5cu1aJFi5SamqpDhw7pvvvuc3vhAABIkpcrnVesWOH0eN68eQoKCtKWLVvUrVs35ebmas6cOVqwYIF69uwpSZo7d65atmyptLQ0derUyX2VAwCgCl4jy83NlSQFBgZKkrZs2aLCwkL17t3b0adFixZq0KCBNm7cWOoYBQUFysvLc1oAALha5Q6y4uJiPf300+rcubPatGkjScrOzpa3t7cCAgKc+gYHBys7O7vUcZKTk+Xv7+9YwsPDy1sSAOAmVO4gi4+P108//aSFCxdWqIDExETl5uY6lqysrAqNBwC4ubh0jazEmDFj9OWXX2rdunWqX7++oz0kJETnzp3TiRMnnI7KcnJyFBISUupYdrtddru9PGUAAODaEZkxRmPGjNHixYv1zTffqHHjxk7rIyMjVa1aNa1Zs8bRlpGRoYMHDyoqKso9FQMAcAGXjsji4+O1YMECLVmyRL6+vo7rXv7+/qpevbr8/f01cuRIJSQkKDAwUH5+fho7dqyioqK4YxEAUClcCrKUlBRJUo8ePZza586dq+HDh0uSpk2bJg8PD8XGxqqgoEDR0dGaNWuWW4oFAOBiLgWZMeaKfXx8fDRz5kzNnDmz3EUBAHC1+KxFAIClEWQAAEsjyAAAlkaQAQAsjSADAFgaQQYAsDSCDABgaQQZAMDSCDIAgKURZAAASyPIAACWRpABACyNIAMAWBpBBgCwNIIMAGBpBBkAwNIIMgCApRFkAABLI8gAAJZGkAEALI0gAwBYGkEGALA0ggwAYGkEGQDA0ggyAICluRxk69at08CBAxUWFiabzabPP//caf3w4cNls9mcln79+rmrXgAAnLgcZPn5+WrXrp1mzpxZZp9+/frp8OHDjuWjjz6qUJEAAJTFy9UNYmJiFBMTc9k+drtdISEhVzVeQUGBCgoKHI/z8vJcLQkAcBOrlGtka9euVVBQkG677TY9/vjjOn78eJl9k5OT5e/v71jCw8MroyQAwA3K7UHWr18//etf/9KaNWv0+uuvKzU1VTExMSoqKiq1f2JionJzcx1LVlaWu0sCANzAXD61eCUPPvig4+e2bdsqIiJCTZs21dq1a9WrV69L+tvtdtntdneXAQC4SVT67fdNmjRR3bp1tWfPnsreFQDgJlTpQfbbb7/p+PHjCg0NrexdAQBuQi6fWjx16pTT0dW+ffuUnp6uwMBABQYGauLEiYqNjVVISIgyMzP1zDPPqFmzZoqOjnZr4QAASOUIsh9++EF3332343FCQoIkKS4uTikpKdq+fbs++OADnThxQmFhYerbt69efvllroMBACqFy0HWo0cPGWPKXL9y5coKFQQAgCv4rEUAgKURZAAASyPIAACWRpABACyNIAMAWBpBBgCwNIIMAGBpBBkAwNIIMgCApRFkAABLI8gAAJZGkAEALI0gAwBYGkEGALA0ggwAYGkEGQDA0ggyAIClEWQAAEsjyAAAlkaQAQAsjSADAFgaQQYAsDSCDABgaQQZAMDSXA6ydevWaeDAgQoLC5PNZtPnn3/utN4Yo5deekmhoaGqXr26evfurd27d7urXgAAnLgcZPn5+WrXrp1mzpxZ6vopU6boH//4h95++21t2rRJNWvWVHR0tM6ePVvhYgEAuJiXqxvExMQoJiam1HXGGE2fPl0vvPCC7r33XknSv/71LwUHB+vzzz/Xgw8+WLFqAQC4iFuvke3bt0/Z2dnq3bu3o83f318dO3bUxo0bS92moKBAeXl5TgsAAFfLrUGWnZ0tSQoODnZqDw4Odqy7WHJysvz9/R1LeHi4O0sCANzgqvyuxcTEROXm5jqWrKysqi4JAGAhbg2ykJAQSVJOTo5Te05OjmPdxex2u/z8/JwWAACulluDrHHjxgoJCdGaNWscbXl5edq0aZOioqLcuSsAACSV467FU6dOac+ePY7H+/btU3p6ugIDA9WgQQM9/fTTeuWVV9S8eXM1btxYL774osLCwjRo0CB31g0AgKRyBNkPP/ygu+++2/E4ISFBkhQXF6d58+bpmWeeUX5+vkaNGqUTJ06oS5cuWrFihXx8fNxXNQAA/8dmjDFVXcSF8vLy5O/vr9zc3ApfL2v07DI3VQUAcMX+yQMqPMbV5kGV37UIAEBFEGQAAEsjyAAAlkaQAQAsjSADAFgaQQYAsDSCDABgaQQZAMDSCDIAgKURZAAASyPIAACWRpABACyNIAMAWBpBBgCwNIIMAGBpBBkAwNIIMgCApRFkAABLI8gAAJZGkAEALI0gAwBYGkEGALA0ggwAYGkEGQDA0ggyAICluT3IJkyYIJvN5rS0aNHC3bsBAECS5FUZg7Zu3VqrV6/+z068KmU3AABUTpB5eXkpJCSkMoYGAMBJpVwj2717t8LCwtSkSRMNHTpUBw8eLLNvQUGB8vLynBYAAK6W24OsY8eOmjdvnlasWKGUlBTt27dPXbt21cmTJ0vtn5ycLH9/f8cSHh7u7pIAADcwmzHGVOYOTpw4oYYNG2rq1KkaOXLkJesLCgpUUFDgeJyXl6fw8HDl5ubKz8+vQvtu9OyyCm0PACif/ZMHVHiMvLw8+fv7XzEPKv0ujICAAN16663as2dPqevtdrvsdntllwEAuEFV+vvITp06pczMTIWGhlb2rgAANyG3B9nf/vY3paamav/+/fruu+/017/+VZ6ennrooYfcvSsAANx/avG3337TQw89pOPHj6tevXrq0qWL0tLSVK9ePXfvCgAA9wfZwoUL3T0kAABl4rMWAQCWRpABACyNIAMAWBpBBgCwNIIMAGBpBBkAwNIIMgCApRFkAABLI8gAAJZGkAEALI0gAwBYGkEGALA0ggwAYGkEGQDA0ggyAIClEWQAAEsjyAAAlkaQAQAsjSADAFgaQQYAsDSCDABgaQQZAMDSCDIAgKURZAAASyPIAACWVmlBNnPmTDVq1Eg+Pj7q2LGjvv/++8raFQDgJlYpQfbxxx8rISFBSUlJ2rp1q9q1a6fo6GgdOXKkMnYHALiJVUqQTZ06VY8++qhGjBihVq1a6e2331aNGjX0/vvvV8buAAA3MS93D3ju3Dlt2bJFiYmJjjYPDw/17t1bGzduvKR/QUGBCgoKHI9zc3MlSXl5eRWupbjgdIXHAAC4zh2v4SVjGGMu28/tQXbs2DEVFRUpODjYqT04OFi//PLLJf2Tk5M1ceLES9rDw8PdXRoA4Brxn+6+sU6ePCl/f/8y17s9yFyVmJiohIQEx+Pi4mL9/vvvqlOnjmw2W7nHzcvLU3h4uLKysuTn5+eOUm8YzE3pmJeyMTelY17K5o65Mcbo5MmTCgsLu2w/twdZ3bp15enpqZycHKf2nJwchYSEXNLfbrfLbrc7tQUEBLitHj8/P/7AysDclI55KRtzUzrmpWwVnZvLHYmVcPvNHt7e3oqMjNSaNWscbcXFxVqzZo2ioqLcvTsAwE2uUk4tJiQkKC4uTnfccYc6dOig6dOnKz8/XyNGjKiM3QEAbmKVEmSDBw/W0aNH9dJLLyk7O1vt27fXihUrLrkBpDLZ7XYlJSVdctoSzE1ZmJeyMTelY17Kdi3nxmaudF8jAADXMT5rEQBgaQQZAMDSCDIAgKURZAAAS7N0kLn6VTGLFi1SixYt5OPjo7Zt22r58uXXqNJry5V5ee+999S1a1fVrl1btWvXVu/evW/or9wp79cLLVy4UDabTYMGDarcAquQq3Nz4sQJxcfHKzQ0VHa7XbfeeusN+W/K1XmZPn26brvtNlWvXl3h4eEaN26czp49e42qvTbWrVungQMHKiwsTDabTZ9//vkVt1m7dq1uv/122e12NWvWTPPmzXNfQcaiFi5caLy9vc37779vfv75Z/Poo4+agIAAk5OTU2r/DRs2GE9PTzNlyhSzc+dO88ILL5hq1aqZHTt2XOPKK5er8zJkyBAzc+ZMs23bNrNr1y4zfPhw4+/vb3777bdrXHnlc3VuSuzbt8/ccsstpmvXrubee++9NsVeY67OTUFBgbnjjjtM//79zfr1682+ffvM2rVrTXp6+jWuvHK5Oi/z5883drvdzJ8/3+zbt8+sXLnShIaGmnHjxl3jyivX8uXLzfPPP28+++wzI8ksXrz4sv337t1ratSoYRISEszOnTvNjBkzjKenp1mxYoVb6rFskHXo0MHEx8c7HhcVFZmwsDCTnJxcav8HHnjADBgwwKmtY8eOZvTo0ZVa57Xm6rxc7Pz588bX19d88MEHlVVilSnP3Jw/f97cddddZvbs2SYuLu6GDTJX5yYlJcU0adLEnDt37lqVWCVcnZf4+HjTs2dPp7aEhATTuXPnSq2zKl1NkD3zzDOmdevWTm2DBw820dHRbqnBkqcWS74qpnfv3o62y31VjCRt3LjRqb8kRUdHl9nfisozLxc7ffq0CgsLFRgYWFllVonyzs2kSZMUFBSkkSNHXosyq0R55uaLL75QVFSU4uPjFRwcrDZt2ui1115TUVHRtSq70pVnXu666y5t2bLFcfpx7969Wr58ufr3739Nar5eVfbrb5V/+n15uPpVMZKUnZ1dav/s7OxKq/NaK8+8XGz8+PEKCwu75I/O6sozN+vXr9ecOXOUnp5+DSqsOuWZm7179+qbb77R0KFDtXz5cu3Zs0dPPPGECgsLlZSUdC3KrnTlmZchQ4bo2LFj6tKli4wxOn/+vB577DE999xz16Lk61ZZr795eXk6c+aMqlevXqHxLXlEhsoxefJkLVy4UIsXL5aPj09Vl1OlTp48qYcffljvvfee6tatW9XlXHeKi4sVFBSkd999V5GRkRo8eLCef/55vf3221VdWpVau3atXnvtNc2aNUtbt27VZ599pmXLlunll1+u6tJuaJY8InP1q2IkKSQkxKX+VlSeeSnx5ptvavLkyVq9erUiIiIqs8wq4ercZGZmav/+/Ro4cKCjrbi4WJLk5eWljIwMNW3atHKLvkbK83cTGhqqatWqydPT09HWsmVLZWdn69y5c/L29q7Umq+F8szLiy++qIcffliPPPKIJKlt27bKz8/XqFGj9Pzzz8vD4+Y8dijr9dfPz6/CR2OSRY/IyvNVMVFRUU79JWnVqlU31FfLlPcrdKZMmaKXX35ZK1as0B133HEtSr3mXJ2bFi1aaMeOHUpPT3cs99xzj+6++26lp6ffUN9gXp6/m86dO2vPnj2OcJekX3/9VaGhoTdEiEnlm5fTp09fElYlYW9u4o+1rfTXX7fcMlIFFi5caOx2u5k3b57ZuXOnGTVqlAkICDDZ2dnGGGMefvhh8+yzzzr6b9iwwXh5eZk333zT7Nq1yyQlJd2wt9+7Mi+TJ0823t7e5tNPPzWHDx92LCdPnqyqp1BpXJ2bi93Idy26OjcHDx40vr6+ZsyYMSYjI8N8+eWXJigoyLzyyitV9RQqhavzkpSUZHx9fc1HH31k9u7da77++mvTtGlT88ADD1TVU6gUJ0+eNNu2bTPbtm0zkszUqVPNtm3bzIEDB4wxxjz77LPm4YcfdvQvuf3+73//u9m1a5eZOXMmt9+XmDFjhmnQoIHx9vY2HTp0MGlpaY513bt3N3FxcU79P/nkE3Prrbcab29v07p1a7Ns2bJrXPG14cq8NGzY0Ei6ZElKSrr2hV8Drv7NXOhGDjJjXJ+b7777znTs2NHY7XbTpEkT8+qrr5rz589f46ornyvzUlhYaCZMmGCaNm1qfHx8THh4uHniiSfMH3/8ce0Lr0Tffvttqa8bJXMRFxdnunfvfsk27du3N97e3qZJkyZm7ty5bquHr3EBAFiaJa+RAQBQgiADAFgaQQYAsDSCDABgaQQZAMDSCDIAgKURZAAASyPIAACWRpABFnC1XycP3IwIMgCApRFkAABLI8iACurRo4fGjBmjMWPGyN/fX3Xr1tWLL74oY4yee+45dezY8ZJt2rVrp0mTJkmSNm/erD59+qhu3bry9/dX9+7dtXXr1jL3t3btWtlsNp04ccLRlp6eLpvNpv379zva1q9fr65du6p69eoKDw/Xk08+qfz8fLc9b+B6QZABbvDBBx/Iy8tL33//vd566y1NnTpVs2fP1tChQ/X9998rMzPT0ffnn3/W9u3bNWTIEEl/fht1XFyc1q9fr7S0NDVv3lz9+/fXyZMny11PZmam+vXrp9jYWG3fvl0ff/yx1q9frzFjxlT4uQLXHbd9jj5wk+revbtp2bKlKS4udrSNHz/etGzZ0hhjTLt27cykSZMc6xITE03Hjh3LHK+oqMj4+vqapUuXOtokmcWLFxtj/vMVGhd+NUjJ90Lt27fPGGPMyJEjzahRo5zG/fe//208PDzMmTNnyvtUgesSR2SAG3Tq1Ek2m83xOCoqSrt371ZRUZGGDh2qBQsWSPrzW4I/+ugjDR061NE3JydHjz76qJo3by5/f3/5+fnp1KlTOnjwYLnr+fHHHzVv3jzVqlXLsURHR6u4uFj79u0r/xMFrkNeVV0AcKN76KGHNH78eG3dulVnzpxRVlaWBg8e7FgfFxen48eP66233lLDhg1lt9sVFRWlc+fOlTqeh8ef//80F3yVYGFhoVOfU6dOafTo0XryyScv2b5BgwbueFrAdYMgA9xg06ZNTo9LrnV5enqqfv366t69u+bPn68zZ86oT58+CgoKcvTdsGGDZs2apf79+0uSsrKydOzYsTL3Va9ePUnS4cOHVbt2bUl/3uxxodtvv107d+5Us2bN3PH0gOsapxYBNzh48KASEhKUkZGhjz76SDNmzNBTTz3lWD906FAtXLhQixYtcjqtKEnNmzfXhx9+qF27dmnTpk0aOnSoqlevXua+mjVrpvDwcE2YMEG7d+/WsmXL9D//8z9OfcaPH6/vvvtOY8aMUXp6unbv3q0lS5ZwswduSAQZ4AbDhg3TmTNn1KFDB8XHx+upp57SqFGjHOv/+7//W8ePH9fp06c1aNAgp23nzJmjP/74Q7fffrsefvhhPfnkk05HbBerVq2aPvroI/3yyy+KiIjQ66+/rldeecWpT0REhFJTU/Xrr7+qa9eu+q//+i+99NJLCgsLc+vzBq4HNnPhiXYALuvRo4fat2+v6dOnV3UpwE2JIzIAgKURZAAAS+PUIgDA0jgiAwBYGkEGALA0ggwAYGkEGQDA0ggyAIClEWQAAEsjyAAAlkaQAQAs7f8DuxESkpssDLgAAAAASUVORK5CYII=\n", + "image/png": "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", "text/plain": [ "
" ] @@ -1359,17 +1370,17 @@ }, { "cell_type": "code", - "execution_count": 18, + "execution_count": 26, "id": "f343d6ea-48be-42fc-ac59-591c0c4afc26", "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "KstestResult(statistic=0.04319182166835675, pvalue=0.6145727388836066, statistic_location=0.08347484499830993, statistic_sign=1)" + "KstestResult(statistic=0.043191821668346606, pvalue=0.6145727388839017, statistic_location=0.08347484499832007, statistic_sign=1)" ] }, - "execution_count": 18, + "execution_count": 26, "metadata": {}, "output_type": "execute_result" } @@ -1409,13 +1420,13 @@ }, { "cell_type": "code", - "execution_count": 19, + "execution_count": 29, "id": "ee4ac1a5-46c7-435b-881a-55c0c94ae419", "metadata": {}, "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "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", "text/plain": [ "
" ] @@ -1431,8 +1442,8 @@ " for _ in range(100)\n", "]\n", "plt.hist(ratio)\n", - "plt.xlabel(\"ratio of man\")\n", - "plt.ylabel(\"probablity\")\n", + "plt.xlabel(\"ratio of man\") #[nits] 英語として正しくない気がする。 male ratio とかでは?\n", + "plt.ylabel(\"probablity\") #[must] 縦軸はprobabilityではない気がする \n", "plt.show()" ] }, @@ -1474,7 +1485,7 @@ }, { "cell_type": "code", - "execution_count": 20, + "execution_count": 30, "id": "e64be9b9-3a93-4030-a26b-067065d32db2", "metadata": {}, "outputs": [], @@ -1505,7 +1516,7 @@ }, { "cell_type": "code", - "execution_count": 21, + "execution_count": 32, "id": "7421cd43-11d1-429b-acbc-cb065bbd4d7f", "metadata": {}, "outputs": [ @@ -1636,7 +1647,7 @@ "[891 rows x 4 columns]" ] }, - "execution_count": 21, + "execution_count": 32, "metadata": {}, "output_type": "execute_result" } @@ -1647,7 +1658,7 @@ }, { "cell_type": "code", - "execution_count": 22, + "execution_count": 33, "id": "50f14069-7522-487d-80e3-539cc9a72675", "metadata": {}, "outputs": [ @@ -1664,7 +1675,7 @@ "Name: is_treat, dtype: float64" ] }, - "execution_count": 22, + "execution_count": 33, "metadata": {}, "output_type": "execute_result" } @@ -1700,7 +1711,7 @@ }, { "cell_type": "code", - "execution_count": 23, + "execution_count": 34, "id": "c9bd39f9-b5e4-447d-ac83-5ebca708290d", "metadata": {}, "outputs": [ @@ -1735,7 +1746,7 @@ "" ] }, - "execution_count": 23, + "execution_count": 34, "metadata": {}, "output_type": "execute_result" } @@ -1749,7 +1760,7 @@ }, { "cell_type": "code", - "execution_count": 24, + "execution_count": 35, "id": "d7fc7055-4427-4c49-abce-889c8e9822e9", "metadata": {}, "outputs": [ @@ -1868,7 +1879,7 @@ "[109 rows x 3 columns]" ] }, - "execution_count": 24, + "execution_count": 35, "metadata": {}, "output_type": "execute_result" } @@ -1893,7 +1904,7 @@ }, { "cell_type": "code", - "execution_count": 25, + "execution_count": 36, "id": "52ee7418-aa7e-4072-88ea-2c48a8db0e10", "metadata": {}, "outputs": [ @@ -1932,7 +1943,7 @@ "" ] }, - "execution_count": 25, + "execution_count": 36, "metadata": {}, "output_type": "execute_result" } @@ -1964,7 +1975,7 @@ }, { "cell_type": "code", - "execution_count": 26, + "execution_count": 37, "id": "7cbf28c7-6366-4c24-9342-d5fd8e3a2464", "metadata": {}, "outputs": [ @@ -1999,7 +2010,7 @@ "" ] }, - "execution_count": 26, + "execution_count": 37, "metadata": {}, "output_type": "execute_result" } @@ -2015,7 +2026,7 @@ }, { "cell_type": "code", - "execution_count": 27, + "execution_count": 38, "id": "a8e34717-2a57-4574-a313-e71c028c2da3", "metadata": {}, "outputs": [ @@ -2146,7 +2157,7 @@ "[1000 rows x 4 columns]" ] }, - "execution_count": 27, + "execution_count": 38, "metadata": {}, "output_type": "execute_result" } @@ -2165,7 +2176,7 @@ }, { "cell_type": "code", - "execution_count": 28, + "execution_count": 39, "id": "da99c6c8-b74a-4677-945d-f3941afffe81", "metadata": {}, "outputs": [ @@ -2178,7 +2189,7 @@ "Name: is_deliver, dtype: float64" ] }, - "execution_count": 28, + "execution_count": 39, "metadata": {}, "output_type": "execute_result" } @@ -2209,7 +2220,7 @@ }, { "cell_type": "code", - "execution_count": 29, + "execution_count": 40, "id": "cdc7816c-3934-4177-8646-9cbc243c99a0", "metadata": {}, "outputs": [ @@ -2250,7 +2261,7 @@ "" ] }, - "execution_count": 29, + "execution_count": 40, "metadata": {}, "output_type": "execute_result" } @@ -2274,7 +2285,7 @@ }, { "cell_type": "code", - "execution_count": 30, + "execution_count": 41, "id": "be3495bd-b845-4a40-9f8f-2590bfcbcde2", "metadata": {}, "outputs": [ @@ -2325,7 +2336,7 @@ "\n", " (20.214) \n", "\n", - "

T-stats reported in parentheses
T-stats use same covariance type as original model
id: 0x13078b940" + "

T-stats reported in parentheses
T-stats use same covariance type as original model
id: 0x16a4c5310" ], "text/plain": [ " First Stage Estimation Results \n", @@ -2349,10 +2360,10 @@ "\n", "T-stats reported in parentheses\n", "T-stats use same covariance type as original model\n", - "FirstStageResults, id: 0x13078b940" + "FirstStageResults, id: 0x16a4c5310" ] }, - "execution_count": 30, + "execution_count": 41, "metadata": {}, "output_type": "execute_result" } @@ -2381,7 +2392,7 @@ }, { "cell_type": "code", - "execution_count": 31, + "execution_count": 43, "id": "ec06f40e-4c4d-4130-87f6-c68af35ffd2e", "metadata": {}, "outputs": [ @@ -2428,7 +2439,7 @@ "" ] }, - "execution_count": 31, + "execution_count": 43, "metadata": {}, "output_type": "execute_result" } @@ -2444,7 +2455,7 @@ }, { "cell_type": "code", - "execution_count": 32, + "execution_count": 44, "id": "0106cda5-06f1-4411-9210-9c55e06c100b", "metadata": {}, "outputs": [ @@ -2587,7 +2598,7 @@ "[50000 rows x 5 columns]" ] }, - "execution_count": 32, + "execution_count": 44, "metadata": {}, "output_type": "execute_result" } @@ -2606,7 +2617,7 @@ }, { "cell_type": "code", - "execution_count": 33, + "execution_count": 45, "id": "02402a2f-3dc7-4fc4-a591-0cdd98408fb5", "metadata": {}, "outputs": [ @@ -2641,7 +2652,7 @@ "" ] }, - "execution_count": 33, + "execution_count": 45, "metadata": {}, "output_type": "execute_result" } @@ -2694,7 +2705,7 @@ }, { "cell_type": "code", - "execution_count": 34, + "execution_count": 47, "id": "6854dd78-ce22-4119-b3f0-3584edeb79ed", "metadata": {}, "outputs": [ @@ -2769,7 +2780,7 @@ "\"\"\"" ] }, - "execution_count": 34, + "execution_count": 47, "metadata": {}, "output_type": "execute_result" } @@ -2797,7 +2808,7 @@ }, { "cell_type": "code", - "execution_count": 35, + "execution_count": 48, "id": "80cdeeae-07bc-42ca-8197-5cfcccfe1784", "metadata": {}, "outputs": [ @@ -2832,7 +2843,7 @@ "" ] }, - "execution_count": 35, + "execution_count": 48, "metadata": {}, "output_type": "execute_result" } @@ -2843,7 +2854,7 @@ }, { "cell_type": "code", - "execution_count": 36, + "execution_count": 49, "id": "826bda17-7a7a-4e77-ab92-15422a17b17a", "metadata": {}, "outputs": [ @@ -2878,7 +2889,7 @@ "" ] }, - "execution_count": 36, + "execution_count": 49, "metadata": {}, "output_type": "execute_result" } @@ -2889,7 +2900,7 @@ }, { "cell_type": "code", - "execution_count": 37, + "execution_count": 50, "id": "3866e741-7850-43a8-9b6e-1a311ba0fa4d", "metadata": {}, "outputs": [ @@ -2984,7 +2995,7 @@ "[50000 rows x 1 columns]" ] }, - "execution_count": 37, + "execution_count": 50, "metadata": {}, "output_type": "execute_result" } @@ -3015,7 +3026,7 @@ }, { "cell_type": "code", - "execution_count": 38, + "execution_count": 51, "id": "6a6b2e1f-9128-4ee2-929a-30fd4a3fb4d2", "metadata": {}, "outputs": [ @@ -3058,7 +3069,7 @@ "" ] }, - "execution_count": 38, + "execution_count": 51, "metadata": {}, "output_type": "execute_result" } @@ -3099,7 +3110,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.8.5" + "version": "3.9.6" } }, "nbformat": 4, diff --git a/notebooks/chapter4_did.ipynb b/notebooks/chapter4_did.ipynb index 55fe12d..8de9df1 100644 --- a/notebooks/chapter4_did.ipynb +++ b/notebooks/chapter4_did.ipynb @@ -18,32 +18,29 @@ "name": "stdout", "output_type": "stream", "text": [ - "Requirement already satisfied: linearmodels in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (4.31)\n", - "Requirement already satisfied: property-cached>=1.6.3 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from linearmodels) (1.6.4)\n", - "Requirement already satisfied: statsmodels>=0.11 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from linearmodels) (0.14.1)\n", - "Requirement already satisfied: setuptools-scm[toml]>=7.0.0 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from linearmodels) (7.1.0)\n", - "Requirement already satisfied: pyhdfe>=0.1 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from linearmodels) (0.2.0)\n", - "Requirement already satisfied: numpy>=1.16 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from linearmodels) (1.24.4)\n", - "Requirement already satisfied: Cython>=0.29.21 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from linearmodels) (3.0.2)\n", - "Requirement already satisfied: scipy>=1.2 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from linearmodels) (1.10.1)\n", - "Requirement already satisfied: pandas>=0.24 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from linearmodels) (2.0.3)\n", - "Requirement already satisfied: mypy-extensions>=0.4 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from linearmodels) (1.0.0)\n", - "Requirement already satisfied: formulaic>=0.3.2 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from linearmodels) (0.6.4)\n", - "Requirement already satisfied: graphlib-backport>=1.0.0 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from formulaic>=0.3.2->linearmodels) (1.0.3)\n", - "Requirement already satisfied: astor>=0.8 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from formulaic>=0.3.2->linearmodels) (0.8.1)\n", - "Requirement already satisfied: interface-meta>=1.2.0 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from formulaic>=0.3.2->linearmodels) (1.3.0)\n", - "Requirement already satisfied: typing-extensions>=4.2.0 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from formulaic>=0.3.2->linearmodels) (4.8.0)\n", - "Requirement already satisfied: wrapt>=1.0 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from formulaic>=0.3.2->linearmodels) (1.15.0)\n", - "Requirement already satisfied: tzdata>=2022.1 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from pandas>=0.24->linearmodels) (2023.4)\n", - "Requirement already satisfied: pytz>=2020.1 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from pandas>=0.24->linearmodels) (2023.3.post1)\n", - "Requirement already satisfied: python-dateutil>=2.8.2 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from pandas>=0.24->linearmodels) (2.8.2)\n", - "Requirement already satisfied: six>=1.5 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from python-dateutil>=2.8.2->pandas>=0.24->linearmodels) (1.16.0)\n", - "Requirement already satisfied: packaging>=20.0 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from setuptools-scm[toml]>=7.0.0->linearmodels) (23.1)\n", - "Requirement already satisfied: setuptools in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from setuptools-scm[toml]>=7.0.0->linearmodels) (52.0.0)\n", - "Requirement already satisfied: tomli>=1.0.0 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from setuptools-scm[toml]>=7.0.0->linearmodels) (2.0.1)\n", - "Requirement already satisfied: patsy>=0.5.4 in /Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages (from statsmodels>=0.11->linearmodels) (0.5.6)\n", - "\u001b[33mWARNING: You are using pip version 21.0.1; however, version 24.0 is available.\n", - "You should consider upgrading via the '/Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/bin/python -m pip install --upgrade pip' command.\u001b[0m\n" + "Requirement already satisfied: linearmodels in /Users/s01057/mylocal/lib/python3.9/site-packages (5.4)\n", + "Requirement already satisfied: statsmodels>=0.12.0 in /Users/s01057/mylocal/lib/python3.9/site-packages (from linearmodels) (0.14.1)\n", + "Requirement already satisfied: pandas>=1.3.0 in /Users/s01057/mylocal/lib/python3.9/site-packages (from linearmodels) (2.1.4)\n", + "Requirement already satisfied: formulaic>=0.6.5 in /Users/s01057/mylocal/lib/python3.9/site-packages (from linearmodels) (1.0.1)\n", + "Requirement already satisfied: numpy>=1.22.0 in /Users/s01057/mylocal/lib/python3.9/site-packages (from linearmodels) (1.26.1)\n", + "Requirement already satisfied: setuptools-scm[toml]<9.0.0,>=8.0.0 in /Users/s01057/mylocal/lib/python3.9/site-packages (from linearmodels) (8.0.4)\n", + "Requirement already satisfied: scipy>=1.5.0 in /Users/s01057/mylocal/lib/python3.9/site-packages (from linearmodels) (1.11.3)\n", + "Requirement already satisfied: mypy-extensions>=0.4 in /Users/s01057/mylocal/lib/python3.9/site-packages (from linearmodels) (1.0.0)\n", + "Requirement already satisfied: Cython>=0.29.37 in /Users/s01057/mylocal/lib/python3.9/site-packages (from linearmodels) (3.0.8)\n", + "Requirement already satisfied: pyhdfe>=0.1 in /Users/s01057/mylocal/lib/python3.9/site-packages (from linearmodels) (0.2.0)\n", + "Requirement already satisfied: interface-meta>=1.2.0 in /Users/s01057/mylocal/lib/python3.9/site-packages (from formulaic>=0.6.5->linearmodels) (1.3.0)\n", + "Requirement already satisfied: typing-extensions>=4.2.0 in /Users/s01057/mylocal/lib/python3.9/site-packages (from formulaic>=0.6.5->linearmodels) (4.8.0)\n", + "Requirement already satisfied: wrapt>=1.0 in /Users/s01057/mylocal/lib/python3.9/site-packages (from formulaic>=0.6.5->linearmodels) (1.16.0)\n", + "Requirement already satisfied: pytz>=2020.1 in /Users/s01057/mylocal/lib/python3.9/site-packages (from pandas>=1.3.0->linearmodels) (2023.3.post1)\n", + "Requirement already satisfied: python-dateutil>=2.8.2 in /Users/s01057/mylocal/lib/python3.9/site-packages (from pandas>=1.3.0->linearmodels) (2.8.2)\n", + "Requirement already satisfied: tzdata>=2022.1 in /Users/s01057/mylocal/lib/python3.9/site-packages (from pandas>=1.3.0->linearmodels) (2023.3)\n", + "Requirement already satisfied: setuptools in /Users/s01057/mylocal/lib/python3.9/site-packages (from setuptools-scm[toml]<9.0.0,>=8.0.0->linearmodels) (58.0.4)\n", + "Requirement already satisfied: packaging>=20 in /Users/s01057/mylocal/lib/python3.9/site-packages (from setuptools-scm[toml]<9.0.0,>=8.0.0->linearmodels) (23.2)\n", + "Requirement already satisfied: tomli>=1 in /Users/s01057/mylocal/lib/python3.9/site-packages (from setuptools-scm[toml]<9.0.0,>=8.0.0->linearmodels) (2.0.1)\n", + "Requirement already satisfied: patsy>=0.5.4 in /Users/s01057/mylocal/lib/python3.9/site-packages (from statsmodels>=0.12.0->linearmodels) (0.5.6)\n", + "Requirement already satisfied: six in /Users/s01057/mylocal/lib/python3.9/site-packages (from patsy>=0.5.4->statsmodels>=0.12.0->linearmodels) (1.16.0)\n", + "\u001b[33mWARNING: There was an error checking the latest version of pip.\u001b[0m\u001b[33m\n", + "\u001b[0m" ] } ], @@ -346,7 +343,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "/Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages/statsmodels/base/model.py:1896: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 28, but rank is 1\n", + "/Users/s01057/mylocal/lib/python3.9/site-packages/statsmodels/base/model.py:1896: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 28, but rank is 1\n", " warnings.warn('covariance of constraints does not have full '\n" ] }, @@ -361,82 +358,82 @@ " Intercept 0.7696 0.004 216.842 0.000 0.762 0.777\n", "\n", "\n", - " C(State)[T.Arizona] -0.5368 8.3e-16 -6.47e+14 0.000 -0.537 -0.537\n", + " C(State)[T.Arizona] -0.5368 1.76e-15 -3.05e+14 0.000 -0.537 -0.537\n", "\n", "\n", " C(State)[T.California] -0.4953 0.004 -139.552 0.000 -0.503 -0.488\n", "\n", "\n", - " C(State)[T.Colorado] -0.1024 1e-15 -1.02e+14 0.000 -0.102 -0.102\n", + " C(State)[T.Colorado] -0.1025 6.28e-16 -1.63e+14 0.000 -0.102 -0.102\n", "\n", "\n", - " C(State)[T.Connecticut] -0.3758 9.61e-16 -3.91e+14 0.000 -0.376 -0.376\n", + " C(State)[T.Connecticut] -0.3759 1.22e-15 -3.09e+14 0.000 -0.376 -0.376\n", "\n", "\n", - " C(State)[T.District of Columbia] -0.4262 8.3e-16 -5.13e+14 0.000 -0.426 -0.426\n", + " C(State)[T.District of Columbia] -0.4263 1.3e-15 -3.28e+14 0.000 -0.426 -0.426\n", "\n", "\n", - " C(State)[T.Florida] -0.3686 9.42e-16 -3.91e+14 0.000 -0.369 -0.369\n", + " C(State)[T.Florida] -0.3686 6.18e-16 -5.96e+14 0.000 -0.369 -0.369\n", "\n", "\n", - " C(State)[T.Hawaii] -0.3467 8.25e-16 -4.2e+14 0.000 -0.347 -0.347\n", + " C(State)[T.Hawaii] -0.3467 1.12e-15 -3.11e+14 0.000 -0.347 -0.347\n", "\n", "\n", - " C(State)[T.Louisiana] -0.1966 8.59e-16 -2.29e+14 0.000 -0.197 -0.197\n", + " C(State)[T.Louisiana] -0.1966 1.89e-15 -1.04e+14 0.000 -0.197 -0.197\n", "\n", "\n", - " C(State)[T.Maryland] -0.3115 1.08e-15 -2.89e+14 0.000 -0.312 -0.312\n", + " C(State)[T.Maryland] -0.3116 1.17e-15 -2.66e+14 0.000 -0.312 -0.312\n", "\n", "\n", - " C(State)[T.Michigan] -0.5558 1.06e-15 -5.24e+14 0.000 -0.556 -0.556\n", + " C(State)[T.Michigan] -0.5558 1.54e-15 -3.62e+14 0.000 -0.556 -0.556\n", "\n", "\n", - " C(State)[T.Minnesota] -0.2526 8.87e-16 -2.85e+14 0.000 -0.253 -0.253\n", + " C(State)[T.Minnesota] -0.2526 1.29e-15 -1.96e+14 0.000 -0.253 -0.253\n", "\n", "\n", - " C(State)[T.Missouri] -0.3674 9.4e-16 -3.91e+14 0.000 -0.367 -0.367\n", + " C(State)[T.Missouri] -0.3674 9.06e-16 -4.06e+14 0.000 -0.367 -0.367\n", "\n", "\n", - " C(State)[T.Montana] -0.1119 1.17e-15 -9.61e+13 0.000 -0.112 -0.112\n", + " C(State)[T.Montana] -0.1119 1.82e-15 -6.14e+13 0.000 -0.112 -0.112\n", "\n", "\n", - " C(State)[T.Nebraska] -0.3367 8.73e-16 -3.86e+14 0.000 -0.337 -0.337\n", + " C(State)[T.Nebraska] -0.3368 5.94e-16 -5.67e+14 0.000 -0.337 -0.337\n", "\n", "\n", - " C(State)[T.New Hampshire] -0.2261 9.17e-16 -2.47e+14 0.000 -0.226 -0.226\n", + " C(State)[T.New Hampshire] -0.2262 7.81e-16 -2.9e+14 0.000 -0.226 -0.226\n", "\n", "\n", - " C(State)[T.New Jersey] -0.4676 1.16e-15 -4.01e+14 0.000 -0.468 -0.468\n", + " C(State)[T.New Jersey] -0.4676 1.29e-15 -3.62e+14 0.000 -0.468 -0.468\n", "\n", "\n", - " C(State)[T.New York] -0.6368 1.3e-15 -4.89e+14 0.000 -0.637 -0.637\n", + " C(State)[T.New York] -0.6368 5.83e-16 -1.09e+15 0.000 -0.637 -0.637\n", "\n", "\n", - " C(State)[T.North Carolina] -0.2420 8.88e-16 -2.72e+14 0.000 -0.242 -0.242\n", + " C(State)[T.North Carolina] -0.2420 9.45e-16 -2.56e+14 0.000 -0.242 -0.242\n", "\n", "\n", - " C(State)[T.Ohio] -0.2049 1.05e-15 -1.94e+14 0.000 -0.205 -0.205\n", + " C(State)[T.Ohio] -0.2049 6.28e-16 -3.26e+14 0.000 -0.205 -0.205\n", "\n", "\n", - " C(State)[T.Pennsylvania] -0.3194 8.6e-16 -3.71e+14 0.000 -0.319 -0.319\n", + " C(State)[T.Pennsylvania] -0.3194 1.22e-15 -2.61e+14 0.000 -0.319 -0.319\n", "\n", "\n", - " C(State)[T.South Carolina] -0.5151 8.73e-16 -5.9e+14 0.000 -0.515 -0.515\n", + " C(State)[T.South Carolina] -0.5152 6.48e-16 -7.95e+14 0.000 -0.515 -0.515\n", "\n", "\n", - " C(State)[T.Tennessee] -0.4345 8.33e-16 -5.21e+14 0.000 -0.435 -0.434\n", + " C(State)[T.Tennessee] -0.4345 9.77e-16 -4.45e+14 0.000 -0.435 -0.434\n", "\n", "\n", - " C(State)[T.Virginia] -0.4227 9.41e-16 -4.49e+14 0.000 -0.423 -0.423\n", + " C(State)[T.Virginia] -0.4228 8.46e-16 -4.99e+14 0.000 -0.423 -0.423\n", "\n", "\n", - " C(State)[T.Washington] -0.1886 9.59e-16 -1.97e+14 0.000 -0.189 -0.189\n", + " C(State)[T.Washington] -0.1886 1.22e-15 -1.54e+14 0.000 -0.189 -0.189\n", "\n", "\n", - " C(State)[T.Wisconsin] -0.2046 9.59e-16 -2.13e+14 0.000 -0.205 -0.205\n", + " C(State)[T.Wisconsin] -0.2046 6.01e-16 -3.4e+14 0.000 -0.205 -0.205\n", "\n", "\n", - " C(State)[T.Wyoming] -0.1826 8.28e-16 -2.21e+14 0.000 -0.183 -0.183\n", + " C(State)[T.Wyoming] -0.1826 8.83e-16 -2.07e+14 0.000 -0.183 -0.183\n", "\n", "\n", " C(Quarter_Num)[T.4] 0.0109 0.007 1.531 0.138 -0.004 0.025\n", @@ -453,32 +450,32 @@ " & \\textbf{coef} & \\textbf{std err} & \\textbf{t} & \\textbf{P$> |$t$|$} & \\textbf{[0.025} & \\textbf{0.975]} \\\\\n", "\\midrule\n", "\\textbf{Intercept} & 0.7696 & 0.004 & 216.842 & 0.000 & 0.762 & 0.777 \\\\\n", - "\\textbf{C(State)[T.Arizona]} & -0.5368 & 8.3e-16 & -6.47e+14 & 0.000 & -0.537 & -0.537 \\\\\n", + "\\textbf{C(State)[T.Arizona]} & -0.5368 & 1.76e-15 & -3.05e+14 & 0.000 & -0.537 & -0.537 \\\\\n", "\\textbf{C(State)[T.California]} & -0.4953 & 0.004 & -139.552 & 0.000 & -0.503 & -0.488 \\\\\n", - "\\textbf{C(State)[T.Colorado]} & -0.1024 & 1e-15 & -1.02e+14 & 0.000 & -0.102 & -0.102 \\\\\n", - "\\textbf{C(State)[T.Connecticut]} & -0.3758 & 9.61e-16 & -3.91e+14 & 0.000 & -0.376 & -0.376 \\\\\n", - "\\textbf{C(State)[T.District of Columbia]} & -0.4262 & 8.3e-16 & -5.13e+14 & 0.000 & -0.426 & -0.426 \\\\\n", - "\\textbf{C(State)[T.Florida]} & -0.3686 & 9.42e-16 & -3.91e+14 & 0.000 & -0.369 & -0.369 \\\\\n", - "\\textbf{C(State)[T.Hawaii]} & -0.3467 & 8.25e-16 & -4.2e+14 & 0.000 & -0.347 & -0.347 \\\\\n", - "\\textbf{C(State)[T.Louisiana]} & -0.1966 & 8.59e-16 & -2.29e+14 & 0.000 & -0.197 & -0.197 \\\\\n", - "\\textbf{C(State)[T.Maryland]} & -0.3115 & 1.08e-15 & -2.89e+14 & 0.000 & -0.312 & -0.312 \\\\\n", - "\\textbf{C(State)[T.Michigan]} & -0.5558 & 1.06e-15 & -5.24e+14 & 0.000 & -0.556 & -0.556 \\\\\n", - "\\textbf{C(State)[T.Minnesota]} & -0.2526 & 8.87e-16 & -2.85e+14 & 0.000 & -0.253 & -0.253 \\\\\n", - "\\textbf{C(State)[T.Missouri]} & -0.3674 & 9.4e-16 & -3.91e+14 & 0.000 & -0.367 & -0.367 \\\\\n", - "\\textbf{C(State)[T.Montana]} & -0.1119 & 1.17e-15 & -9.61e+13 & 0.000 & -0.112 & -0.112 \\\\\n", - "\\textbf{C(State)[T.Nebraska]} & -0.3367 & 8.73e-16 & -3.86e+14 & 0.000 & -0.337 & -0.337 \\\\\n", - "\\textbf{C(State)[T.New Hampshire]} & -0.2261 & 9.17e-16 & -2.47e+14 & 0.000 & -0.226 & -0.226 \\\\\n", - "\\textbf{C(State)[T.New Jersey]} & -0.4676 & 1.16e-15 & -4.01e+14 & 0.000 & -0.468 & -0.468 \\\\\n", - "\\textbf{C(State)[T.New York]} & -0.6368 & 1.3e-15 & -4.89e+14 & 0.000 & -0.637 & -0.637 \\\\\n", - "\\textbf{C(State)[T.North Carolina]} & -0.2420 & 8.88e-16 & -2.72e+14 & 0.000 & -0.242 & -0.242 \\\\\n", - "\\textbf{C(State)[T.Ohio]} & -0.2049 & 1.05e-15 & -1.94e+14 & 0.000 & -0.205 & -0.205 \\\\\n", - "\\textbf{C(State)[T.Pennsylvania]} & -0.3194 & 8.6e-16 & -3.71e+14 & 0.000 & -0.319 & -0.319 \\\\\n", - "\\textbf{C(State)[T.South Carolina]} & -0.5151 & 8.73e-16 & -5.9e+14 & 0.000 & -0.515 & -0.515 \\\\\n", - "\\textbf{C(State)[T.Tennessee]} & -0.4345 & 8.33e-16 & -5.21e+14 & 0.000 & -0.435 & -0.434 \\\\\n", - "\\textbf{C(State)[T.Virginia]} & -0.4227 & 9.41e-16 & -4.49e+14 & 0.000 & -0.423 & -0.423 \\\\\n", - "\\textbf{C(State)[T.Washington]} & -0.1886 & 9.59e-16 & -1.97e+14 & 0.000 & -0.189 & -0.189 \\\\\n", - "\\textbf{C(State)[T.Wisconsin]} & -0.2046 & 9.59e-16 & -2.13e+14 & 0.000 & -0.205 & -0.205 \\\\\n", - "\\textbf{C(State)[T.Wyoming]} & -0.1826 & 8.28e-16 & -2.21e+14 & 0.000 & -0.183 & -0.183 \\\\\n", + "\\textbf{C(State)[T.Colorado]} & -0.1025 & 6.28e-16 & -1.63e+14 & 0.000 & -0.102 & -0.102 \\\\\n", + "\\textbf{C(State)[T.Connecticut]} & -0.3759 & 1.22e-15 & -3.09e+14 & 0.000 & -0.376 & -0.376 \\\\\n", + "\\textbf{C(State)[T.District of Columbia]} & -0.4263 & 1.3e-15 & -3.28e+14 & 0.000 & -0.426 & -0.426 \\\\\n", + "\\textbf{C(State)[T.Florida]} & -0.3686 & 6.18e-16 & -5.96e+14 & 0.000 & -0.369 & -0.369 \\\\\n", + "\\textbf{C(State)[T.Hawaii]} & -0.3467 & 1.12e-15 & -3.11e+14 & 0.000 & -0.347 & -0.347 \\\\\n", + "\\textbf{C(State)[T.Louisiana]} & -0.1966 & 1.89e-15 & -1.04e+14 & 0.000 & -0.197 & -0.197 \\\\\n", + "\\textbf{C(State)[T.Maryland]} & -0.3116 & 1.17e-15 & -2.66e+14 & 0.000 & -0.312 & -0.312 \\\\\n", + "\\textbf{C(State)[T.Michigan]} & -0.5558 & 1.54e-15 & -3.62e+14 & 0.000 & -0.556 & -0.556 \\\\\n", + "\\textbf{C(State)[T.Minnesota]} & -0.2526 & 1.29e-15 & -1.96e+14 & 0.000 & -0.253 & -0.253 \\\\\n", + "\\textbf{C(State)[T.Missouri]} & -0.3674 & 9.06e-16 & -4.06e+14 & 0.000 & -0.367 & -0.367 \\\\\n", + "\\textbf{C(State)[T.Montana]} & -0.1119 & 1.82e-15 & -6.14e+13 & 0.000 & -0.112 & -0.112 \\\\\n", + "\\textbf{C(State)[T.Nebraska]} & -0.3368 & 5.94e-16 & -5.67e+14 & 0.000 & -0.337 & -0.337 \\\\\n", + "\\textbf{C(State)[T.New Hampshire]} & -0.2262 & 7.81e-16 & -2.9e+14 & 0.000 & -0.226 & -0.226 \\\\\n", + "\\textbf{C(State)[T.New Jersey]} & -0.4676 & 1.29e-15 & -3.62e+14 & 0.000 & -0.468 & -0.468 \\\\\n", + "\\textbf{C(State)[T.New York]} & -0.6368 & 5.83e-16 & -1.09e+15 & 0.000 & -0.637 & -0.637 \\\\\n", + "\\textbf{C(State)[T.North Carolina]} & -0.2420 & 9.45e-16 & -2.56e+14 & 0.000 & -0.242 & -0.242 \\\\\n", + "\\textbf{C(State)[T.Ohio]} & -0.2049 & 6.28e-16 & -3.26e+14 & 0.000 & -0.205 & -0.205 \\\\\n", + "\\textbf{C(State)[T.Pennsylvania]} & -0.3194 & 1.22e-15 & -2.61e+14 & 0.000 & -0.319 & -0.319 \\\\\n", + "\\textbf{C(State)[T.South Carolina]} & -0.5152 & 6.48e-16 & -7.95e+14 & 0.000 & -0.515 & -0.515 \\\\\n", + "\\textbf{C(State)[T.Tennessee]} & -0.4345 & 9.77e-16 & -4.45e+14 & 0.000 & -0.435 & -0.434 \\\\\n", + "\\textbf{C(State)[T.Virginia]} & -0.4228 & 8.46e-16 & -4.99e+14 & 0.000 & -0.423 & -0.423 \\\\\n", + "\\textbf{C(State)[T.Washington]} & -0.1886 & 1.22e-15 & -1.54e+14 & 0.000 & -0.189 & -0.189 \\\\\n", + "\\textbf{C(State)[T.Wisconsin]} & -0.2046 & 6.01e-16 & -3.4e+14 & 0.000 & -0.205 & -0.205 \\\\\n", + "\\textbf{C(State)[T.Wyoming]} & -0.1826 & 8.83e-16 & -2.07e+14 & 0.000 & -0.183 & -0.183 \\\\\n", "\\textbf{C(Quarter\\_Num)[T.4]} & 0.0109 & 0.007 & 1.531 & 0.138 & -0.004 & 0.025 \\\\\n", "\\textbf{IsTreatment} & -0.0216 & 0.007 & -3.038 & 0.005 & -0.036 & -0.007 \\\\\n", "\\bottomrule\n", @@ -535,16 +532,16 @@ " Intercept 0.7696 0.004 216.842 0.000 0.762 0.777\n", "\n", "\n", - " C(State)[T.Arizona] -0.5368 8.3e-16 -6.47e+14 0.000 -0.537 -0.537\n", + " C(State)[T.Arizona] -0.5368 1.76e-15 -3.05e+14 0.000 -0.537 -0.537\n", "\n", "\n", " C(State)[T.California] -0.4953 0.004 -139.552 0.000 -0.503 -0.488\n", "\n", "\n", - " C(State)[T.Wisconsin] -0.2046 9.59e-16 -2.13e+14 0.000 -0.205 -0.205\n", + " C(State)[T.Wisconsin] -0.2046 6.01e-16 -3.4e+14 0.000 -0.205 -0.205\n", "\n", "\n", - " C(State)[T.Wyoming] -0.1826 8.28e-16 -2.21e+14 0.000 -0.183 -0.183\n", + " C(State)[T.Wyoming] -0.1826 8.83e-16 -2.07e+14 0.000 -0.183 -0.183\n", "\n", "\n", " C(Quarter_Num)[T.4] 0.0109 0.007 1.531 0.138 -0.004 0.025\n", @@ -561,10 +558,10 @@ " & \\textbf{coef} & \\textbf{std err} & \\textbf{t} & \\textbf{P$> |$t$|$} & \\textbf{[0.025} & \\textbf{0.975]} \\\\\n", "\\midrule\n", "\\textbf{Intercept} & 0.7696 & 0.004 & 216.842 & 0.000 & 0.762 & 0.777 \\\\\n", - "\\textbf{C(State)[T.Arizona]} & -0.5368 & 8.3e-16 & -6.47e+14 & 0.000 & -0.537 & -0.537 \\\\\n", + "\\textbf{C(State)[T.Arizona]} & -0.5368 & 1.76e-15 & -3.05e+14 & 0.000 & -0.537 & -0.537 \\\\\n", "\\textbf{C(State)[T.California]} & -0.4953 & 0.004 & -139.552 & 0.000 & -0.503 & -0.488 \\\\\n", - "\\textbf{C(State)[T.Wisconsin]} & -0.2046 & 9.59e-16 & -2.13e+14 & 0.000 & -0.205 & -0.205 \\\\\n", - "\\textbf{C(State)[T.Wyoming]} & -0.1826 & 8.28e-16 & -2.21e+14 & 0.000 & -0.183 & -0.183 \\\\\n", + "\\textbf{C(State)[T.Wisconsin]} & -0.2046 & 6.01e-16 & -3.4e+14 & 0.000 & -0.205 & -0.205 \\\\\n", + "\\textbf{C(State)[T.Wyoming]} & -0.1826 & 8.83e-16 & -2.07e+14 & 0.000 & -0.183 & -0.183 \\\\\n", "\\textbf{C(Quarter\\_Num)[T.4]} & 0.0109 & 0.007 & 1.531 & 0.138 & -0.004 & 0.025 \\\\\n", "\\textbf{IsTreatment} & -0.0216 & 0.007 & -3.038 & 0.005 & -0.036 & -0.007 \\\\\n", "\\bottomrule\n", @@ -679,7 +676,7 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 12, "id": "e2c94517-376b-4532-b021-58f6f9aa26ea", "metadata": {}, "outputs": [ @@ -687,7 +684,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "/Users/s13592/Documents/project/intro_to_impact_evaluation_with_python/.venv/lib/python3.8/site-packages/statsmodels/base/model.py:1896: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 32, but rank is 5\n", + "/Users/s01057/mylocal/lib/python3.9/site-packages/statsmodels/base/model.py:1896: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 32, but rank is 5\n", " warnings.warn('covariance of constraints does not have full '\n" ] }, @@ -702,82 +699,82 @@ " Intercept 0.7655 0.005 147.502 0.000 0.755 0.776\n", "\n", "\n", - " C(State)[T.Arizona] -0.5329 8.85e-16 -6.02e+14 0.000 -0.533 -0.533\n", + " C(State)[T.Arizona] -0.5329 6.28e-15 -8.48e+13 0.000 -0.533 -0.533\n", "\n", "\n", " C(State)[T.California] -0.4950 0.003 -147.316 0.000 -0.502 -0.488\n", "\n", "\n", - " C(State)[T.Colorado] -0.1055 8.59e-16 -1.23e+14 0.000 -0.105 -0.105\n", + " C(State)[T.Colorado] -0.1055 1.41e-15 -7.46e+13 0.000 -0.105 -0.105\n", "\n", "\n", - " C(State)[T.Connecticut] -0.3802 1.33e-15 -2.87e+14 0.000 -0.380 -0.380\n", + " C(State)[T.Connecticut] -0.3802 3.88e-15 -9.8e+13 0.000 -0.380 -0.380\n", "\n", "\n", - " C(State)[T.District of Columbia] -0.4301 1.04e-15 -4.15e+14 0.000 -0.430 -0.430\n", + " C(State)[T.District of Columbia] -0.4300 1.06e-15 -4.06e+14 0.000 -0.430 -0.430\n", "\n", "\n", - " C(State)[T.Florida] -0.3770 8.62e-16 -4.37e+14 0.000 -0.377 -0.377\n", + " C(State)[T.Florida] -0.3770 1.12e-15 -3.37e+14 0.000 -0.377 -0.377\n", "\n", "\n", - " C(State)[T.Hawaii] -0.3524 8.32e-16 -4.23e+14 0.000 -0.352 -0.352\n", + " C(State)[T.Hawaii] -0.3524 1.76e-16 -2.01e+15 0.000 -0.352 -0.352\n", "\n", "\n", - " C(State)[T.Louisiana] -0.2167 1.06e-15 -2.04e+14 0.000 -0.217 -0.217\n", + " C(State)[T.Louisiana] -0.2167 8.56e-16 -2.53e+14 0.000 -0.217 -0.217\n", "\n", "\n", - " C(State)[T.Maryland] -0.3076 8.62e-16 -3.57e+14 0.000 -0.308 -0.308\n", + " C(State)[T.Maryland] -0.3076 7.52e-16 -4.09e+14 0.000 -0.308 -0.308\n", "\n", "\n", - " C(State)[T.Michigan] -0.5409 1.39e-15 -3.89e+14 0.000 -0.541 -0.541\n", + " C(State)[T.Michigan] -0.5409 3.66e-15 -1.48e+14 0.000 -0.541 -0.541\n", "\n", "\n", - " C(State)[T.Minnesota] -0.2457 9.08e-16 -2.7e+14 0.000 -0.246 -0.246\n", + " C(State)[T.Minnesota] -0.2457 1.14e-15 -2.16e+14 0.000 -0.246 -0.246\n", "\n", "\n", - " C(State)[T.Missouri] -0.3660 1.2e-15 -3.04e+14 0.000 -0.366 -0.366\n", + " C(State)[T.Missouri] -0.3660 2.33e-15 -1.57e+14 0.000 -0.366 -0.366\n", "\n", "\n", - " C(State)[T.Montana] -0.1185 8.43e-16 -1.41e+14 0.000 -0.118 -0.118\n", + " C(State)[T.Montana] -0.1184 2.35e-16 -5.05e+14 0.000 -0.118 -0.118\n", "\n", "\n", - " C(State)[T.Nebraska] -0.3308 8.15e-16 -4.06e+14 0.000 -0.331 -0.331\n", + " C(State)[T.Nebraska] -0.3308 7.92e-16 -4.18e+14 0.000 -0.331 -0.331\n", "\n", "\n", - " C(State)[T.New Hampshire] -0.2292 8.03e-16 -2.85e+14 0.000 -0.229 -0.229\n", + " C(State)[T.New Hampshire] -0.2292 4.77e-17 -4.8e+15 0.000 -0.229 -0.229\n", "\n", "\n", - " C(State)[T.New Jersey] -0.4607 1.16e-15 -3.99e+14 0.000 -0.461 -0.461\n", + " C(State)[T.New Jersey] -0.4607 2e-16 -2.3e+15 0.000 -0.461 -0.461\n", "\n", "\n", - " C(State)[T.New York] -0.6436 8.04e-16 -8.01e+14 0.000 -0.644 -0.644\n", + " C(State)[T.New York] -0.6436 1.19e-15 -5.43e+14 0.000 -0.644 -0.644\n", "\n", "\n", - " C(State)[T.North Carolina] -0.2484 8.32e-16 -2.99e+14 0.000 -0.248 -0.248\n", + " C(State)[T.North Carolina] -0.2484 5.66e-16 -4.39e+14 0.000 -0.248 -0.248\n", "\n", "\n", - " C(State)[T.Ohio] -0.2077 1.17e-15 -1.78e+14 0.000 -0.208 -0.208\n", + " C(State)[T.Ohio] -0.2077 4.87e-16 -4.27e+14 0.000 -0.208 -0.208\n", "\n", "\n", - " C(State)[T.Pennsylvania] -0.3191 1.02e-15 -3.12e+14 0.000 -0.319 -0.319\n", + " C(State)[T.Pennsylvania] -0.3190 4.44e-16 -7.19e+14 0.000 -0.319 -0.319\n", "\n", "\n", - " C(State)[T.South Carolina] -0.5043 9.57e-16 -5.27e+14 0.000 -0.504 -0.504\n", + " C(State)[T.South Carolina] -0.5043 1.21e-15 -4.15e+14 0.000 -0.504 -0.504\n", "\n", "\n", - " C(State)[T.Tennessee] -0.4381 1.5e-15 -2.92e+14 0.000 -0.438 -0.438\n", + " C(State)[T.Tennessee] -0.4381 6.1e-16 -7.18e+14 0.000 -0.438 -0.438\n", "\n", "\n", - " C(State)[T.Virginia] -0.4324 1.07e-15 -4.04e+14 0.000 -0.432 -0.432\n", + " C(State)[T.Virginia] -0.4324 4.37e-17 -9.89e+15 0.000 -0.432 -0.432\n", "\n", "\n", - " C(State)[T.Washington] -0.1863 9.62e-16 -1.94e+14 0.000 -0.186 -0.186\n", + " C(State)[T.Washington] -0.1863 4.85e-16 -3.84e+14 0.000 -0.186 -0.186\n", "\n", "\n", - " C(State)[T.Wisconsin] -0.2012 1.11e-15 -1.82e+14 0.000 -0.201 -0.201\n", + " C(State)[T.Wisconsin] -0.2012 4.44e-16 -4.53e+14 0.000 -0.201 -0.201\n", "\n", "\n", - " C(State)[T.Wyoming] -0.1823 8.99e-16 -2.03e+14 0.000 -0.182 -0.182\n", + " C(State)[T.Wyoming] -0.1823 7.38e-16 -2.47e+14 0.000 -0.182 -0.182\n", "\n", "\n", " C(Quarter_Num)[T.2] -0.0024 0.006 -0.420 0.678 -0.014 0.009\n", @@ -806,32 +803,32 @@ " & \\textbf{coef} & \\textbf{std err} & \\textbf{t} & \\textbf{P$> |$t$|$} & \\textbf{[0.025} & \\textbf{0.975]} \\\\\n", "\\midrule\n", "\\textbf{Intercept} & 0.7655 & 0.005 & 147.502 & 0.000 & 0.755 & 0.776 \\\\\n", - "\\textbf{C(State)[T.Arizona]} & -0.5329 & 8.85e-16 & -6.02e+14 & 0.000 & -0.533 & -0.533 \\\\\n", + "\\textbf{C(State)[T.Arizona]} & -0.5329 & 6.28e-15 & -8.48e+13 & 0.000 & -0.533 & -0.533 \\\\\n", "\\textbf{C(State)[T.California]} & -0.4950 & 0.003 & -147.316 & 0.000 & -0.502 & -0.488 \\\\\n", - "\\textbf{C(State)[T.Colorado]} & -0.1055 & 8.59e-16 & -1.23e+14 & 0.000 & -0.105 & -0.105 \\\\\n", - "\\textbf{C(State)[T.Connecticut]} & -0.3802 & 1.33e-15 & -2.87e+14 & 0.000 & -0.380 & -0.380 \\\\\n", - "\\textbf{C(State)[T.District of Columbia]} & -0.4301 & 1.04e-15 & -4.15e+14 & 0.000 & -0.430 & -0.430 \\\\\n", - "\\textbf{C(State)[T.Florida]} & -0.3770 & 8.62e-16 & -4.37e+14 & 0.000 & -0.377 & -0.377 \\\\\n", - "\\textbf{C(State)[T.Hawaii]} & -0.3524 & 8.32e-16 & -4.23e+14 & 0.000 & -0.352 & -0.352 \\\\\n", - "\\textbf{C(State)[T.Louisiana]} & -0.2167 & 1.06e-15 & -2.04e+14 & 0.000 & -0.217 & -0.217 \\\\\n", - "\\textbf{C(State)[T.Maryland]} & -0.3076 & 8.62e-16 & -3.57e+14 & 0.000 & -0.308 & -0.308 \\\\\n", - "\\textbf{C(State)[T.Michigan]} & -0.5409 & 1.39e-15 & -3.89e+14 & 0.000 & -0.541 & -0.541 \\\\\n", - "\\textbf{C(State)[T.Minnesota]} & -0.2457 & 9.08e-16 & -2.7e+14 & 0.000 & -0.246 & -0.246 \\\\\n", - "\\textbf{C(State)[T.Missouri]} & -0.3660 & 1.2e-15 & -3.04e+14 & 0.000 & -0.366 & -0.366 \\\\\n", - "\\textbf{C(State)[T.Montana]} & -0.1185 & 8.43e-16 & -1.41e+14 & 0.000 & -0.118 & -0.118 \\\\\n", - "\\textbf{C(State)[T.Nebraska]} & -0.3308 & 8.15e-16 & -4.06e+14 & 0.000 & -0.331 & -0.331 \\\\\n", - "\\textbf{C(State)[T.New Hampshire]} & -0.2292 & 8.03e-16 & -2.85e+14 & 0.000 & -0.229 & -0.229 \\\\\n", - "\\textbf{C(State)[T.New Jersey]} & -0.4607 & 1.16e-15 & -3.99e+14 & 0.000 & -0.461 & -0.461 \\\\\n", - "\\textbf{C(State)[T.New York]} & -0.6436 & 8.04e-16 & -8.01e+14 & 0.000 & -0.644 & -0.644 \\\\\n", - "\\textbf{C(State)[T.North Carolina]} & -0.2484 & 8.32e-16 & -2.99e+14 & 0.000 & -0.248 & -0.248 \\\\\n", - "\\textbf{C(State)[T.Ohio]} & -0.2077 & 1.17e-15 & -1.78e+14 & 0.000 & -0.208 & -0.208 \\\\\n", - "\\textbf{C(State)[T.Pennsylvania]} & -0.3191 & 1.02e-15 & -3.12e+14 & 0.000 & -0.319 & -0.319 \\\\\n", - "\\textbf{C(State)[T.South Carolina]} & -0.5043 & 9.57e-16 & -5.27e+14 & 0.000 & -0.504 & -0.504 \\\\\n", - "\\textbf{C(State)[T.Tennessee]} & -0.4381 & 1.5e-15 & -2.92e+14 & 0.000 & -0.438 & -0.438 \\\\\n", - "\\textbf{C(State)[T.Virginia]} & -0.4324 & 1.07e-15 & -4.04e+14 & 0.000 & -0.432 & -0.432 \\\\\n", - "\\textbf{C(State)[T.Washington]} & -0.1863 & 9.62e-16 & -1.94e+14 & 0.000 & -0.186 & -0.186 \\\\\n", - "\\textbf{C(State)[T.Wisconsin]} & -0.2012 & 1.11e-15 & -1.82e+14 & 0.000 & -0.201 & -0.201 \\\\\n", - "\\textbf{C(State)[T.Wyoming]} & -0.1823 & 8.99e-16 & -2.03e+14 & 0.000 & -0.182 & -0.182 \\\\\n", + "\\textbf{C(State)[T.Colorado]} & -0.1055 & 1.41e-15 & -7.46e+13 & 0.000 & -0.105 & -0.105 \\\\\n", + "\\textbf{C(State)[T.Connecticut]} & -0.3802 & 3.88e-15 & -9.8e+13 & 0.000 & -0.380 & -0.380 \\\\\n", + "\\textbf{C(State)[T.District of Columbia]} & -0.4300 & 1.06e-15 & -4.06e+14 & 0.000 & -0.430 & -0.430 \\\\\n", + "\\textbf{C(State)[T.Florida]} & -0.3770 & 1.12e-15 & -3.37e+14 & 0.000 & -0.377 & -0.377 \\\\\n", + "\\textbf{C(State)[T.Hawaii]} & -0.3524 & 1.76e-16 & -2.01e+15 & 0.000 & -0.352 & -0.352 \\\\\n", + "\\textbf{C(State)[T.Louisiana]} & -0.2167 & 8.56e-16 & -2.53e+14 & 0.000 & -0.217 & -0.217 \\\\\n", + "\\textbf{C(State)[T.Maryland]} & -0.3076 & 7.52e-16 & -4.09e+14 & 0.000 & -0.308 & -0.308 \\\\\n", + "\\textbf{C(State)[T.Michigan]} & -0.5409 & 3.66e-15 & -1.48e+14 & 0.000 & -0.541 & -0.541 \\\\\n", + "\\textbf{C(State)[T.Minnesota]} & -0.2457 & 1.14e-15 & -2.16e+14 & 0.000 & -0.246 & -0.246 \\\\\n", + "\\textbf{C(State)[T.Missouri]} & -0.3660 & 2.33e-15 & -1.57e+14 & 0.000 & -0.366 & -0.366 \\\\\n", + "\\textbf{C(State)[T.Montana]} & -0.1184 & 2.35e-16 & -5.05e+14 & 0.000 & -0.118 & -0.118 \\\\\n", + "\\textbf{C(State)[T.Nebraska]} & -0.3308 & 7.92e-16 & -4.18e+14 & 0.000 & -0.331 & -0.331 \\\\\n", + "\\textbf{C(State)[T.New Hampshire]} & -0.2292 & 4.77e-17 & -4.8e+15 & 0.000 & -0.229 & -0.229 \\\\\n", + "\\textbf{C(State)[T.New Jersey]} & -0.4607 & 2e-16 & -2.3e+15 & 0.000 & -0.461 & -0.461 \\\\\n", + "\\textbf{C(State)[T.New York]} & -0.6436 & 1.19e-15 & -5.43e+14 & 0.000 & -0.644 & -0.644 \\\\\n", + "\\textbf{C(State)[T.North Carolina]} & -0.2484 & 5.66e-16 & -4.39e+14 & 0.000 & -0.248 & -0.248 \\\\\n", + "\\textbf{C(State)[T.Ohio]} & -0.2077 & 4.87e-16 & -4.27e+14 & 0.000 & -0.208 & -0.208 \\\\\n", + "\\textbf{C(State)[T.Pennsylvania]} & -0.3190 & 4.44e-16 & -7.19e+14 & 0.000 & -0.319 & -0.319 \\\\\n", + "\\textbf{C(State)[T.South Carolina]} & -0.5043 & 1.21e-15 & -4.15e+14 & 0.000 & -0.504 & -0.504 \\\\\n", + "\\textbf{C(State)[T.Tennessee]} & -0.4381 & 6.1e-16 & -7.18e+14 & 0.000 & -0.438 & -0.438 \\\\\n", + "\\textbf{C(State)[T.Virginia]} & -0.4324 & 4.37e-17 & -9.89e+15 & 0.000 & -0.432 & -0.432 \\\\\n", + "\\textbf{C(State)[T.Washington]} & -0.1863 & 4.85e-16 & -3.84e+14 & 0.000 & -0.186 & -0.186 \\\\\n", + "\\textbf{C(State)[T.Wisconsin]} & -0.2012 & 4.44e-16 & -4.53e+14 & 0.000 & -0.201 & -0.201 \\\\\n", + "\\textbf{C(State)[T.Wyoming]} & -0.1823 & 7.38e-16 & -2.47e+14 & 0.000 & -0.182 & -0.182 \\\\\n", "\\textbf{C(Quarter\\_Num)[T.2]} & -0.0024 & 0.006 & -0.420 & 0.678 & -0.014 & 0.009 \\\\\n", "\\textbf{C(Quarter\\_Num)[T.3]} & 0.0049 & 0.005 & 0.919 & 0.367 & -0.006 & 0.016 \\\\\n", "\\textbf{C(Quarter\\_Num)[T.4]} & 0.0158 & 0.007 & 2.217 & 0.036 & 0.001 & 0.030 \\\\\n", @@ -846,7 +843,7 @@ "" ] }, - "execution_count": 9, + "execution_count": 12, "metadata": {}, "output_type": "execute_result" } @@ -860,7 +857,7 @@ " data=df_organ_donations_full,\n", ").fit()\n", "# 標準誤差の補正\n", - "result_correted = result.get_robustcov_results(\n", + "result_correted = result.get_robustcov_results( # [nits] result_corrected? 以下全て同様\n", " \"cluster\", groups=df_organ_donations_full[\"State\"]\n", ")\n", "# 結果の出力\n", @@ -869,7 +866,7 @@ }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 13, "id": "ec4d1647-48cb-4349-8584-3210aa219c1c", "metadata": {}, "outputs": [ @@ -1034,7 +1031,7 @@ "9 0 " ] }, - "execution_count": 10, + "execution_count": 13, "metadata": {}, "output_type": "execute_result" } @@ -1053,7 +1050,7 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 14, "id": "2fdd2277-281b-4192-9c71-f5e9baae5011", "metadata": {}, "outputs": [ @@ -1068,16 +1065,16 @@ " Intercept 0.7655 0.005 147.502 0.000 0.755 0.776\n", "\n", "\n", - " C(State)[T.Arizona] -0.5329 8.85e-16 -6.02e+14 0.000 -0.533 -0.533\n", + " C(State)[T.Arizona] -0.5329 6.28e-15 -8.48e+13 0.000 -0.533 -0.533\n", "\n", "\n", " C(State)[T.California] -0.4950 0.003 -147.316 0.000 -0.502 -0.488\n", "\n", "\n", - " C(State)[T.Wisconsin] -0.2012 1.11e-15 -1.82e+14 0.000 -0.201 -0.201\n", + " C(State)[T.Wisconsin] -0.2012 4.44e-16 -4.53e+14 0.000 -0.201 -0.201\n", "\n", "\n", - " C(State)[T.Wyoming] -0.1823 8.99e-16 -2.03e+14 0.000 -0.182 -0.182\n", + " C(State)[T.Wyoming] -0.1823 7.38e-16 -2.47e+14 0.000 -0.182 -0.182\n", "\n", "\n", " C(Quarter_Num)[T.2] -0.0024 0.006 -0.420 0.678 -0.014 0.009\n", @@ -1106,10 +1103,10 @@ " & \\textbf{coef} & \\textbf{std err} & \\textbf{t} & \\textbf{P$> |$t$|$} & \\textbf{[0.025} & \\textbf{0.975]} \\\\\n", "\\midrule\n", "\\textbf{Intercept} & 0.7655 & 0.005 & 147.502 & 0.000 & 0.755 & 0.776 \\\\\n", - "\\textbf{C(State)[T.Arizona]} & -0.5329 & 8.85e-16 & -6.02e+14 & 0.000 & -0.533 & -0.533 \\\\\n", + "\\textbf{C(State)[T.Arizona]} & -0.5329 & 6.28e-15 & -8.48e+13 & 0.000 & -0.533 & -0.533 \\\\\n", "\\textbf{C(State)[T.California]} & -0.4950 & 0.003 & -147.316 & 0.000 & -0.502 & -0.488 \\\\\n", - "\\textbf{C(State)[T.Wisconsin]} & -0.2012 & 1.11e-15 & -1.82e+14 & 0.000 & -0.201 & -0.201 \\\\\n", - "\\textbf{C(State)[T.Wyoming]} & -0.1823 & 8.99e-16 & -2.03e+14 & 0.000 & -0.182 & -0.182 \\\\\n", + "\\textbf{C(State)[T.Wisconsin]} & -0.2012 & 4.44e-16 & -4.53e+14 & 0.000 & -0.201 & -0.201 \\\\\n", + "\\textbf{C(State)[T.Wyoming]} & -0.1823 & 7.38e-16 & -2.47e+14 & 0.000 & -0.182 & -0.182 \\\\\n", "\\textbf{C(Quarter\\_Num)[T.2]} & -0.0024 & 0.006 & -0.420 & 0.678 & -0.014 & 0.009 \\\\\n", "\\textbf{C(Quarter\\_Num)[T.3]} & 0.0049 & 0.005 & 0.919 & 0.367 & -0.006 & 0.016 \\\\\n", "\\textbf{C(Quarter\\_Num)[T.4]} & 0.0158 & 0.007 & 2.217 & 0.036 & 0.001 & 0.030 \\\\\n", @@ -1124,7 +1121,7 @@ "" ] }, - "execution_count": 11, + "execution_count": 14, "metadata": {}, "output_type": "execute_result" } @@ -1965,7 +1962,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.8.5" + "version": "3.9.6" } }, "nbformat": 4,