{
"cells": [
{
"cell_type": "code",
"execution_count": 24,
"id": "e06644f1-96f9-46ef-9ba0-23346e145442",
"metadata": {},
"outputs": [],
"source": [
"import pandas as pd\n",
"import statsmodels.formula.api as smf\n",
"import numpy as np\n",
"import matplotlib.pyplot as plt"
]
},
{
"cell_type": "markdown",
"id": "fb7b16df-0987-4a58-a76c-7392c2729bea",
"metadata": {},
"source": [
"# What Is Panel Data? "
]
},
{
"cell_type": "markdown",
"id": "0a02ed5a-8026-4e01-a1a9-8f33dbf46a62",
"metadata": {},
"source": [
"**Panel data** is a hybrid data type that has feature of both _cross section_ and _time series_. Actually panel data are the most common data type in industry, for instance a car manufacturer has record of its suppliers' price level over time, a bank has full history of its clients' monthly balance for many years. Needless to say, to carry out serious researches, you must use panel data.\n",
"\n",
"\n",
"Here we will use the data from \"Why has Productivity Declined? Productivity and Public Investment\" written by Munell, A.\n",
"\n",
"Variable names defined as below:\n",
"```\n",
"STATE = state name\n",
"ST_ABB = state abbreviation\n",
"YR = 1970,...,1986\n",
"P_CAP = public capital\n",
"HWY = highway capital\n",
"WATER = water utility capital\n",
"UTIL = utility capital\n",
"PC = private capital\n",
"GSP = gross state product\n",
"EMP = employment\n",
"UNEMP = unemployment rate\n",
"```"
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "a141ba3c-91e5-409f-bda2-65e6190db5d3",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
" \n",
" \n",
" STATE \n",
" YR \n",
" P_CAP \n",
" HWY \n",
" WATER \n",
" UTIL \n",
" PC \n",
" GSP \n",
" EMP \n",
" UNEMP \n",
" \n",
" \n",
" \n",
" \n",
" 0 \n",
" ALABAMA \n",
" 1970 \n",
" 15032.67 \n",
" 7325.80 \n",
" 1655.68 \n",
" 6051.20 \n",
" 35793.80 \n",
" 28418 \n",
" 1010.5 \n",
" 4.7 \n",
" \n",
" \n",
" 1 \n",
" ALABAMA \n",
" 1971 \n",
" 15501.94 \n",
" 7525.94 \n",
" 1721.02 \n",
" 6254.98 \n",
" 37299.91 \n",
" 29375 \n",
" 1021.9 \n",
" 5.2 \n",
" \n",
" \n",
" 2 \n",
" ALABAMA \n",
" 1972 \n",
" 15972.41 \n",
" 7765.42 \n",
" 1764.75 \n",
" 6442.23 \n",
" 38670.30 \n",
" 31303 \n",
" 1072.3 \n",
" 4.7 \n",
" \n",
" \n",
" 3 \n",
" ALABAMA \n",
" 1973 \n",
" 16406.26 \n",
" 7907.66 \n",
" 1742.41 \n",
" 6756.19 \n",
" 40084.01 \n",
" 33430 \n",
" 1135.5 \n",
" 3.9 \n",
" \n",
" \n",
" 4 \n",
" ALABAMA \n",
" 1974 \n",
" 16762.67 \n",
" 8025.52 \n",
" 1734.85 \n",
" 7002.29 \n",
" 42057.31 \n",
" 33749 \n",
" 1169.8 \n",
" 5.5 \n",
" \n",
" \n",
"
\n",
"
"
],
"text/plain": [
" STATE YR P_CAP HWY WATER UTIL PC GSP \\\n",
"0 ALABAMA 1970 15032.67 7325.80 1655.68 6051.20 35793.80 28418 \n",
"1 ALABAMA 1971 15501.94 7525.94 1721.02 6254.98 37299.91 29375 \n",
"2 ALABAMA 1972 15972.41 7765.42 1764.75 6442.23 38670.30 31303 \n",
"3 ALABAMA 1973 16406.26 7907.66 1742.41 6756.19 40084.01 33430 \n",
"4 ALABAMA 1974 16762.67 8025.52 1734.85 7002.29 42057.31 33749 \n",
"\n",
" EMP UNEMP \n",
"0 1010.5 4.7 \n",
"1 1021.9 5.2 \n",
"2 1072.3 4.7 \n",
"3 1135.5 3.9 \n",
"4 1169.8 5.5 "
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df = pd.read_excel(\n",
" \"Basic_Econometrics_practice_data.xlsx\", sheet_name=\"Prod_PubInvestment\"\n",
")\n",
"df.head(5)"
]
},
{
"cell_type": "code",
"execution_count": 9,
"id": "c8b5605d-b401-4bc9-890a-a3f00afd772d",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" \n",
" STATE \n",
" YR \n",
" P_CAP \n",
" HWY \n",
" WATER \n",
" UTIL \n",
" PC \n",
" GSP \n",
" EMP \n",
" UNEMP \n",
" \n",
" \n",
" \n",
" \n",
" 811 \n",
" WYOMING \n",
" 1982 \n",
" 4731.98 \n",
" 3060.64 \n",
" 408.43 \n",
" 1262.90 \n",
" 27724.96 \n",
" 13056 \n",
" 217.7 \n",
" 5.8 \n",
" \n",
" \n",
" 812 \n",
" WYOMING \n",
" 1983 \n",
" 4950.82 \n",
" 3119.98 \n",
" 445.59 \n",
" 1385.25 \n",
" 28586.46 \n",
" 11922 \n",
" 202.5 \n",
" 8.4 \n",
" \n",
" \n",
" 813 \n",
" WYOMING \n",
" 1984 \n",
" 5184.73 \n",
" 3195.68 \n",
" 476.57 \n",
" 1512.48 \n",
" 28794.80 \n",
" 12073 \n",
" 204.3 \n",
" 6.3 \n",
" \n",
" \n",
" 814 \n",
" WYOMING \n",
" 1985 \n",
" 5448.38 \n",
" 3295.92 \n",
" 523.01 \n",
" 1629.45 \n",
" 29326.94 \n",
" 12022 \n",
" 206.9 \n",
" 7.1 \n",
" \n",
" \n",
" 815 \n",
" WYOMING \n",
" 1986 \n",
" 5700.41 \n",
" 3400.96 \n",
" 565.58 \n",
" 1733.88 \n",
" 27110.51 \n",
" 10870 \n",
" 196.3 \n",
" 9 \n",
" \n",
" \n",
"
\n",
"
"
],
"text/plain": [
" STATE YR P_CAP HWY WATER UTIL PC GSP EMP \\\n",
"811 WYOMING 1982 4731.98 3060.64 408.43 1262.90 27724.96 13056 217.7 \n",
"812 WYOMING 1983 4950.82 3119.98 445.59 1385.25 28586.46 11922 202.5 \n",
"813 WYOMING 1984 5184.73 3195.68 476.57 1512.48 28794.80 12073 204.3 \n",
"814 WYOMING 1985 5448.38 3295.92 523.01 1629.45 29326.94 12022 206.9 \n",
"815 WYOMING 1986 5700.41 3400.96 565.58 1733.88 27110.51 10870 196.3 \n",
"\n",
" UNEMP \n",
"811 5.8 \n",
"812 8.4 \n",
"813 6.3 \n",
"814 7.1 \n",
"815 9 "
]
},
"execution_count": 9,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.tail(5)"
]
},
{
"cell_type": "markdown",
"id": "36072c69-8757-4984-8712-3b13221299e1",
"metadata": {},
"source": [
"Each state is recorded over time in several aspects, such as public capitals, highway capital, water facility capital and etc. If each state is recorded in equal length of time period, we call it **balanced panel**, otherwise **unbalanced panel**."
]
},
{
"cell_type": "markdown",
"id": "d0908310-5bde-42e3-aa58-137220ac88c9",
"metadata": {},
"source": [
"Estimation methods includes four approaches\n",
"1. Pooled OLS model\n",
"2. Fixed effects least square dummy variable (LSDV) model\n",
"3. Fixed effects within-in group model\n",
"4. Random effects model"
]
},
{
"cell_type": "markdown",
"id": "e2c7f836-af65-45cb-83b9-58faf37924c3",
"metadata": {},
"source": [
"# Pooled OLS Regression "
]
},
{
"cell_type": "markdown",
"id": "820001b6-e371-4c16-a6cf-667a2d030052",
"metadata": {},
"source": [
"\\begin{aligned}\n",
"ln{GSP}_{i t} &=\\beta_{1}+\\beta_{2} \\ln{PCAP}_{i t}+\\beta_{3} \\ln{HWY}_{i t}+\\beta_{4} \\ln{WATER}_{i t}+\\beta_{5} \\ln{UTIL}_{i t}+\\beta_{6} \\ln{EMP}_{i t}+u_{i t}\n",
"\\end{aligned}\n",
"where $i$ means the $i$the state, $t$ means time period."
]
},
{
"cell_type": "code",
"execution_count": 23,
"id": "b8ed122c-cc95-4b4e-a82c-701e7b3b0a64",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
" OLS Regression Results \n",
"==============================================================================\n",
"Dep. Variable: np.log(GSP) R-squared: 0.993\n",
"Model: OLS Adj. R-squared: 0.993\n",
"Method: Least Squares F-statistic: 1.971e+04\n",
"Date: Mon, 11 Oct 2021 Prob (F-statistic): 0.00\n",
"Time: 22:39:33 Log-Likelihood: 862.28\n",
"No. Observations: 816 AIC: -1711.\n",
"Df Residuals: 809 BIC: -1678.\n",
"Df Model: 6 \n",
"Covariance Type: nonrobust \n",
"=================================================================================\n",
" coef std err t P>|t| [0.025 0.975]\n",
"---------------------------------------------------------------------------------\n",
"Intercept 1.2474 0.111 11.245 0.000 1.030 1.465\n",
"np.log(P_CAP) 0.7432 0.109 6.824 0.000 0.529 0.957\n",
"np.log(PC) 0.3124 0.011 28.553 0.000 0.291 0.334\n",
"np.log(HWY) -0.3289 0.060 -5.520 0.000 -0.446 -0.212\n",
"np.log(WATER) 0.0276 0.018 1.568 0.117 -0.007 0.062\n",
"np.log(UTIL) -0.3036 0.047 -6.510 0.000 -0.395 -0.212\n",
"np.log(EMP) 0.5932 0.016 36.761 0.000 0.561 0.625\n",
"==============================================================================\n",
"Omnibus: 17.950 Durbin-Watson: 0.200\n",
"Prob(Omnibus): 0.000 Jarque-Bera (JB): 19.040\n",
"Skew: 0.328 Prob(JB): 7.34e-05\n",
"Kurtosis: 3.360 Cond. No. 1.23e+03\n",
"==============================================================================\n",
"\n",
"Notes:\n",
"[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
"[2] The condition number is large, 1.23e+03. This might indicate that there are\n",
"strong multicollinearity or other numerical problems.\n"
]
}
],
"source": [
"model = smf.ols(\n",
" formula=\"np.log(GSP) ~ np.log(P_CAP) + np.log(PC) + np.log(HWY) + np.log(WATER) + np.log(UTIL) + np.log(EMP)\",\n",
" data=df,\n",
")\n",
"results = model.fit()\n",
"print(results.summary())"
]
},
{
"cell_type": "markdown",
"id": "4720a771-23cb-4463-9821-f577487ac310",
"metadata": {},
"source": [
"The common symptoms of pooled regression on panel data is that all most of coefficients will be highly significant and also $R^2$ is exceedingly high. However, we can still spot some problems, the conditional number is high, meaning multicollinearity and Durbin-Watson test is close to $0$ meaning autocorrelation or specification error."
]
},
{
"cell_type": "markdown",
"id": "b23077b7-72ad-4160-b4ad-85f3d2c40d9f",
"metadata": {},
"source": [
"But the most prominent issue of this model is that it camouflages the heterogeneity that may exist among states. The heterogeneity of each state is subsumed by the disturbance term, which causes correlation between independent variables and disturbance terms, therefore OLS estimates are bound to be biased and inconsistent."
]
},
{
"cell_type": "markdown",
"id": "78c169eb-c41a-49be-845b-4ffc83eb55dc",
"metadata": {},
"source": [
"# The Fixed Effect LSDV Model "
]
},
{
"cell_type": "markdown",
"id": "3f03cb2e-1c90-448a-af78-e711b7930963",
"metadata": {},
"source": [
"LSDV model allows heterogeneity to take part in by adding different intercept value"
]
},
{
"cell_type": "markdown",
"id": "6e607ccd-5ae9-41ec-ae57-aef27cfd7c4e",
"metadata": {},
"source": [
"\\begin{aligned}\n",
"ln{GSP}_{i t} &=\\beta_{1i}+\\beta_{2} \\ln{PCAP}_{i t}+\\beta_{3} \\ln{HWY}_{i t}+\\beta_{4} \\ln{WATER}_{i t}+\\beta_{5} \\ln{UTIL}_{i t}+\\beta_{6} \\ln{EMP}_{i t}+u_{i t}\n",
"\\end{aligned}"
]
},
{
"cell_type": "markdown",
"id": "0b086044-fe83-42c4-b469-7badf3b491fd",
"metadata": {},
"source": [
"$\\beta_{1i}$ represents the intercept for each state $i$. There are various possible reasons contributing to heterogeneity among states, such as population, average education level and urbanization rate, etc.\n",
"\n",
"_Fixed effect_ means that though each state has its own intercept, but it is **time-invariant**, i.e. constant over the time. If we assume **time-variant** intercept, the notation would be $\\beta_{1it}$"
]
},
{
"cell_type": "code",
"execution_count": 25,
"id": "effe1439-8a83-4f12-8a92-ab1b40df9b51",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" \n",
" STATE \n",
" YR \n",
" P_CAP \n",
" HWY \n",
" WATER \n",
" UTIL \n",
" PC \n",
" GSP \n",
" EMP \n",
" UNEMP \n",
" \n",
" \n",
" \n",
" \n",
" 0 \n",
" ALABAMA \n",
" 1970 \n",
" 15032.67 \n",
" 7325.80 \n",
" 1655.68 \n",
" 6051.20 \n",
" 35793.80 \n",
" 28418 \n",
" 1010.5 \n",
" 4.7 \n",
" \n",
" \n",
" 1 \n",
" ALABAMA \n",
" 1971 \n",
" 15501.94 \n",
" 7525.94 \n",
" 1721.02 \n",
" 6254.98 \n",
" 37299.91 \n",
" 29375 \n",
" 1021.9 \n",
" 5.2 \n",
" \n",
" \n",
" 2 \n",
" ALABAMA \n",
" 1972 \n",
" 15972.41 \n",
" 7765.42 \n",
" 1764.75 \n",
" 6442.23 \n",
" 38670.30 \n",
" 31303 \n",
" 1072.3 \n",
" 4.7 \n",
" \n",
" \n",
" 3 \n",
" ALABAMA \n",
" 1973 \n",
" 16406.26 \n",
" 7907.66 \n",
" 1742.41 \n",
" 6756.19 \n",
" 40084.01 \n",
" 33430 \n",
" 1135.5 \n",
" 3.9 \n",
" \n",
" \n",
" 4 \n",
" ALABAMA \n",
" 1974 \n",
" 16762.67 \n",
" 8025.52 \n",
" 1734.85 \n",
" 7002.29 \n",
" 42057.31 \n",
" 33749 \n",
" 1169.8 \n",
" 5.5 \n",
" \n",
" \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" \n",
" \n",
" 811 \n",
" WYOMING \n",
" 1982 \n",
" 4731.98 \n",
" 3060.64 \n",
" 408.43 \n",
" 1262.90 \n",
" 27724.96 \n",
" 13056 \n",
" 217.7 \n",
" 5.8 \n",
" \n",
" \n",
" 812 \n",
" WYOMING \n",
" 1983 \n",
" 4950.82 \n",
" 3119.98 \n",
" 445.59 \n",
" 1385.25 \n",
" 28586.46 \n",
" 11922 \n",
" 202.5 \n",
" 8.4 \n",
" \n",
" \n",
" 813 \n",
" WYOMING \n",
" 1984 \n",
" 5184.73 \n",
" 3195.68 \n",
" 476.57 \n",
" 1512.48 \n",
" 28794.80 \n",
" 12073 \n",
" 204.3 \n",
" 6.3 \n",
" \n",
" \n",
" 814 \n",
" WYOMING \n",
" 1985 \n",
" 5448.38 \n",
" 3295.92 \n",
" 523.01 \n",
" 1629.45 \n",
" 29326.94 \n",
" 12022 \n",
" 206.9 \n",
" 7.1 \n",
" \n",
" \n",
" 815 \n",
" WYOMING \n",
" 1986 \n",
" 5700.41 \n",
" 3400.96 \n",
" 565.58 \n",
" 1733.88 \n",
" 27110.51 \n",
" 10870 \n",
" 196.3 \n",
" 9 \n",
" \n",
" \n",
"
\n",
"
816 rows × 10 columns
\n",
"
"
],
"text/plain": [
" STATE YR P_CAP HWY WATER UTIL PC GSP \\\n",
"0 ALABAMA 1970 15032.67 7325.80 1655.68 6051.20 35793.80 28418 \n",
"1 ALABAMA 1971 15501.94 7525.94 1721.02 6254.98 37299.91 29375 \n",
"2 ALABAMA 1972 15972.41 7765.42 1764.75 6442.23 38670.30 31303 \n",
"3 ALABAMA 1973 16406.26 7907.66 1742.41 6756.19 40084.01 33430 \n",
"4 ALABAMA 1974 16762.67 8025.52 1734.85 7002.29 42057.31 33749 \n",
".. ... ... ... ... ... ... ... ... \n",
"811 WYOMING 1982 4731.98 3060.64 408.43 1262.90 27724.96 13056 \n",
"812 WYOMING 1983 4950.82 3119.98 445.59 1385.25 28586.46 11922 \n",
"813 WYOMING 1984 5184.73 3195.68 476.57 1512.48 28794.80 12073 \n",
"814 WYOMING 1985 5448.38 3295.92 523.01 1629.45 29326.94 12022 \n",
"815 WYOMING 1986 5700.41 3400.96 565.58 1733.88 27110.51 10870 \n",
"\n",
" EMP UNEMP \n",
"0 1010.5 4.7 \n",
"1 1021.9 5.2 \n",
"2 1072.3 4.7 \n",
"3 1135.5 3.9 \n",
"4 1169.8 5.5 \n",
".. ... ... \n",
"811 217.7 5.8 \n",
"812 202.5 8.4 \n",
"813 204.3 6.3 \n",
"814 206.9 7.1 \n",
"815 196.3 9 \n",
"\n",
"[816 rows x 10 columns]"
]
},
"execution_count": 25,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df"
]
},
{
"cell_type": "code",
"execution_count": 44,
"id": "83cbc246-eb70-4542-a51b-fe4ab08453a0",
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"fig, ax = plt.subplots(nrows=2, ncols=3, figsize=(18, 12))\n",
"ax[0, 0].scatter(df[\"GSP\"], df[\"P_CAP\"], c=\"r\", s=5)\n",
"ax[0, 0].grid()\n",
"ax[0, 0].set_xlabel(\"Public Capital\")\n",
"ax[0, 0].set_ylabel(\"Gross Regional Produce\")\n",
"\n",
"ax[0, 1].scatter(df[\"GSP\"], df[\"HWY\"], c=\"r\", s=5)\n",
"ax[0, 1].grid()\n",
"ax[0, 1].set_xlabel(\"High Way Capital\")\n",
"ax[0, 1].set_ylabel(\"Gross Regional Produce\")\n",
"\n",
"ax[0, 2].scatter(df[\"GSP\"], df[\"WATER\"], c=\"r\", s=5)\n",
"ax[0, 2].grid()\n",
"ax[0, 2].set_xlabel(\"Water Facility\")\n",
"ax[0, 2].set_ylabel(\"Gross Regional Produce\")\n",
"\n",
"ax[1, 0].scatter(df[\"GSP\"], df[\"UTIL\"], c=\"r\", s=5)\n",
"ax[1, 0].grid()\n",
"ax[1, 0].set_xlabel(\"Utiltiy Capital\")\n",
"ax[1, 0].set_ylabel(\"Gross Regional Produce\")\n",
"\n",
"ax[1, 1].scatter(df[\"GSP\"], df[\"PC\"], c=\"r\", s=5)\n",
"ax[1, 1].grid()\n",
"ax[1, 1].set_xlabel(\"Private Capital\")\n",
"ax[1, 1].set_ylabel(\"Gross Regional Produce\")\n",
"\n",
"ax[1, 2].scatter(df[\"GSP\"], df[\"EMP\"], c=\"r\", s=5)\n",
"ax[1, 2].grid()\n",
"ax[1, 2].set_xlabel(\"Employement\")\n",
"ax[1, 2].set_ylabel(\"Gross Regional Produce\")\n",
"\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"id": "40ad195a-56c6-4654-9b74-8d2c47765e91",
"metadata": {},
"source": [
"Check how many states are there in the panel data"
]
},
{
"cell_type": "code",
"execution_count": 59,
"id": "17d43c76-2cf1-45ce-b336-b9a5186dac39",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"['ALABAMA' 'ARIZONA' 'ARKANSAS' 'CALIFORNIA' 'COLORADO' 'CONNECTICUT'\n",
" 'DELAWARE' 'FLORIDA' 'GEORGIA' 'IDAHO' 'ILLINOIS' 'INDIANA' 'IOWA'\n",
" 'KANSAS' 'KENTUCKY' 'LOUISIANA' 'MAINE' 'MARYLAND' 'MASSACHUSETTS'\n",
" 'MICHIGAN' 'MINNESOTA' 'MISSISSIPPI' 'MISSOURI' 'MONTANA' 'NEBRASKA'\n",
" 'NEVADA' 'NEW_HAMPSHIRE' 'NEW_JERSEY' 'NEW_MEXICO' 'NEW_YORK'\n",
" 'NORTH_CAROLINA' 'NORTH_DAKOTA' 'OHIO' 'OKLAHOMA' 'OREGON' 'PENNSYLVANIA'\n",
" 'RHODE_ISLAND' 'SOUTH_CAROLINA' 'SOUTH_DAKOTA' 'TENNESSE' 'TEXAS' 'UTAH'\n",
" 'VERMONT' 'VIRGINIA' 'WASHINGTON' 'WEST_VIRGINIA' 'WISCONSIN' 'WYOMING']\n",
"48\n"
]
}
],
"source": [
"print(df[\"STATE\"].unique())\n",
"print(len(df[\"STATE\"].unique()))"
]
},
{
"cell_type": "markdown",
"id": "7c36c7f0-ae11-468b-a504-013dee943165",
"metadata": {},
"source": [
"To avoid dummy variable trap, we can define $47$ dummy intercepts."
]
},
{
"cell_type": "markdown",
"id": "460c162d-8930-43c2-a66a-52309fa27c35",
"metadata": {},
"source": [
"Add dummies onto the intercept\n",
"\n",
"\\begin{aligned}\n",
"ln{GSP}_{i t} &=\\alpha_{1}+ \\sum_{j=2}^{48}\\alpha_{j} D_{j i}+\\beta_{2} \\ln{PCAP}_{i t}+\\beta_{3} \\ln{HWY}_{i t}+\\beta_{4} \\ln{WATER}_{i t}+\\beta_{5} \\ln{UTIL}_{i t}+\\beta_{6} \\ln{EMP}_{i t}+u_{i t}\n",
"\\end{aligned}\n"
]
},
{
"cell_type": "markdown",
"id": "6746e1b1-476a-466c-a615-45dc2fa832df",
"metadata": {},
"source": [
"Use ```STATE``` as the dummy column and add ```drop_fist``` to avoid dummy trap."
]
},
{
"cell_type": "code",
"execution_count": 62,
"id": "5673cd51-ca67-4bb0-8e9f-99d8fa725deb",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"df_dum = pd.get_dummies(data=df, columns=[\"STATE\"], drop_first=True)"
]
},
{
"cell_type": "code",
"execution_count": 72,
"id": "a647cc53-ab42-4e03-961b-2eba9e3864e8",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" \n",
" YR \n",
" P_CAP \n",
" HWY \n",
" WATER \n",
" UTIL \n",
" PC \n",
" GSP \n",
" EMP \n",
" UNEMP \n",
" STATE_ARIZONA \n",
" ... \n",
" STATE_SOUTH_DAKOTA \n",
" STATE_TENNESSE \n",
" STATE_TEXAS \n",
" STATE_UTAH \n",
" STATE_VERMONT \n",
" STATE_VIRGINIA \n",
" STATE_WASHINGTON \n",
" STATE_WEST_VIRGINIA \n",
" STATE_WISCONSIN \n",
" STATE_WYOMING \n",
" \n",
" \n",
" \n",
" \n",
" 0 \n",
" 1970 \n",
" 15032.67 \n",
" 7325.80 \n",
" 1655.68 \n",
" 6051.20 \n",
" 35793.80 \n",
" 28418 \n",
" 1010.5 \n",
" 4.7 \n",
" 0 \n",
" ... \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" \n",
" \n",
" 1 \n",
" 1971 \n",
" 15501.94 \n",
" 7525.94 \n",
" 1721.02 \n",
" 6254.98 \n",
" 37299.91 \n",
" 29375 \n",
" 1021.9 \n",
" 5.2 \n",
" 0 \n",
" ... \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" \n",
" \n",
" 2 \n",
" 1972 \n",
" 15972.41 \n",
" 7765.42 \n",
" 1764.75 \n",
" 6442.23 \n",
" 38670.30 \n",
" 31303 \n",
" 1072.3 \n",
" 4.7 \n",
" 0 \n",
" ... \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" \n",
" \n",
" 3 \n",
" 1973 \n",
" 16406.26 \n",
" 7907.66 \n",
" 1742.41 \n",
" 6756.19 \n",
" 40084.01 \n",
" 33430 \n",
" 1135.5 \n",
" 3.9 \n",
" 0 \n",
" ... \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" \n",
" \n",
" 4 \n",
" 1974 \n",
" 16762.67 \n",
" 8025.52 \n",
" 1734.85 \n",
" 7002.29 \n",
" 42057.31 \n",
" 33749 \n",
" 1169.8 \n",
" 5.5 \n",
" 0 \n",
" ... \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" \n",
" \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" ... \n",
" \n",
" \n",
" 811 \n",
" 1982 \n",
" 4731.98 \n",
" 3060.64 \n",
" 408.43 \n",
" 1262.90 \n",
" 27724.96 \n",
" 13056 \n",
" 217.7 \n",
" 5.8 \n",
" 0 \n",
" ... \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 1 \n",
" \n",
" \n",
" 812 \n",
" 1983 \n",
" 4950.82 \n",
" 3119.98 \n",
" 445.59 \n",
" 1385.25 \n",
" 28586.46 \n",
" 11922 \n",
" 202.5 \n",
" 8.4 \n",
" 0 \n",
" ... \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 1 \n",
" \n",
" \n",
" 813 \n",
" 1984 \n",
" 5184.73 \n",
" 3195.68 \n",
" 476.57 \n",
" 1512.48 \n",
" 28794.80 \n",
" 12073 \n",
" 204.3 \n",
" 6.3 \n",
" 0 \n",
" ... \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 1 \n",
" \n",
" \n",
" 814 \n",
" 1985 \n",
" 5448.38 \n",
" 3295.92 \n",
" 523.01 \n",
" 1629.45 \n",
" 29326.94 \n",
" 12022 \n",
" 206.9 \n",
" 7.1 \n",
" 0 \n",
" ... \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 1 \n",
" \n",
" \n",
" 815 \n",
" 1986 \n",
" 5700.41 \n",
" 3400.96 \n",
" 565.58 \n",
" 1733.88 \n",
" 27110.51 \n",
" 10870 \n",
" 196.3 \n",
" 9 \n",
" 0 \n",
" ... \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 0 \n",
" 1 \n",
" \n",
" \n",
"
\n",
"
816 rows × 56 columns
\n",
"
"
],
"text/plain": [
" YR P_CAP HWY WATER UTIL PC GSP EMP UNEMP \\\n",
"0 1970 15032.67 7325.80 1655.68 6051.20 35793.80 28418 1010.5 4.7 \n",
"1 1971 15501.94 7525.94 1721.02 6254.98 37299.91 29375 1021.9 5.2 \n",
"2 1972 15972.41 7765.42 1764.75 6442.23 38670.30 31303 1072.3 4.7 \n",
"3 1973 16406.26 7907.66 1742.41 6756.19 40084.01 33430 1135.5 3.9 \n",
"4 1974 16762.67 8025.52 1734.85 7002.29 42057.31 33749 1169.8 5.5 \n",
".. ... ... ... ... ... ... ... ... ... \n",
"811 1982 4731.98 3060.64 408.43 1262.90 27724.96 13056 217.7 5.8 \n",
"812 1983 4950.82 3119.98 445.59 1385.25 28586.46 11922 202.5 8.4 \n",
"813 1984 5184.73 3195.68 476.57 1512.48 28794.80 12073 204.3 6.3 \n",
"814 1985 5448.38 3295.92 523.01 1629.45 29326.94 12022 206.9 7.1 \n",
"815 1986 5700.41 3400.96 565.58 1733.88 27110.51 10870 196.3 9 \n",
"\n",
" STATE_ARIZONA ... STATE_SOUTH_DAKOTA STATE_TENNESSE STATE_TEXAS \\\n",
"0 0 ... 0 0 0 \n",
"1 0 ... 0 0 0 \n",
"2 0 ... 0 0 0 \n",
"3 0 ... 0 0 0 \n",
"4 0 ... 0 0 0 \n",
".. ... ... ... ... ... \n",
"811 0 ... 0 0 0 \n",
"812 0 ... 0 0 0 \n",
"813 0 ... 0 0 0 \n",
"814 0 ... 0 0 0 \n",
"815 0 ... 0 0 0 \n",
"\n",
" STATE_UTAH STATE_VERMONT STATE_VIRGINIA STATE_WASHINGTON \\\n",
"0 0 0 0 0 \n",
"1 0 0 0 0 \n",
"2 0 0 0 0 \n",
"3 0 0 0 0 \n",
"4 0 0 0 0 \n",
".. ... ... ... ... \n",
"811 0 0 0 0 \n",
"812 0 0 0 0 \n",
"813 0 0 0 0 \n",
"814 0 0 0 0 \n",
"815 0 0 0 0 \n",
"\n",
" STATE_WEST_VIRGINIA STATE_WISCONSIN STATE_WYOMING \n",
"0 0 0 0 \n",
"1 0 0 0 \n",
"2 0 0 0 \n",
"3 0 0 0 \n",
"4 0 0 0 \n",
".. ... ... ... \n",
"811 0 0 1 \n",
"812 0 0 1 \n",
"813 0 0 1 \n",
"814 0 0 1 \n",
"815 0 0 1 \n",
"\n",
"[816 rows x 56 columns]"
]
},
"execution_count": 72,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df_dum"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "88526b36-1894-4fb6-810b-b36cbb439dec",
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.8.8"
}
},
"nbformat": 4,
"nbformat_minor": 5
}