Re: 문제21. (점심시간 문제) 미국 의료비 데이터를 다중회귀분석하세요 ~~

[12기_유혜영]

import numpy as np

import statsmodels.api as sm

import statsmodels.formula.api as smf

import pandas as pd

from  sklearn.preprocessing import StandardScaler   # 표준화를 위해 필요

# 1. 데이터 불러오기

df = pd.read_csv("c:\\data\\insurance.csv", engine='python', encoding='CP949')

print(df)

# 2. 모델 생성하기

model = smf.ols(formula = 'expenses ~ age + sex + bmi + children + smoker + region', data = df)

result = model.fit()  # 모델 훈련

print( result.summary() )

결과:

                            OLS Regression Results                           

==============================================================================

Dep. Variable:               expenses   R-squared:                       0.751

Model:                            OLS   Adj. R-squared:                  0.749

Method:                 Least Squares   F-statistic:                     500.9

Date:                Wed, 24 Feb 2021   Prob (F-statistic):               0.00

Time:                        12:07:08   Log-Likelihood:                -13548.

No. Observations:                1338   AIC:                         2.711e+04

Df Residuals:                    1329   BIC:                         2.716e+04

Df Model:                           8                                        

Covariance Type:            nonrobust                                        

=======================================================================================

                          coef    std err          t      P>|t|      [0.025      0.975]

---------------------------------------------------------------------------------------

Intercept           -1.194e+04    987.811    -12.089      0.000   -1.39e+04      -1e+04

sex[T.male]          -131.3520    332.935     -0.395      0.693    -784.488     521.784

smoker[T.yes]        2.385e+04    413.139     57.723      0.000     2.3e+04    2.47e+04

region[T.northwest]  -352.7901    476.261     -0.741      0.459   -1287.095     581.515

region[T.southeast] -1035.5957    478.681     -2.163      0.031   -1974.648     -96.544

region[T.southwest]  -959.3058    477.912     -2.007      0.045   -1896.850     -21.762

age                   256.8392     11.899     21.586      0.000     233.497     280.181

bmi                   339.2899     28.598     11.864      0.000     283.187     395.393

children              475.6889    137.800      3.452      0.001     205.360     746.017

==============================================================================

Omnibus:                      300.499   Durbin-Watson:                   2.088

Prob(Omnibus):                  0.000   Jarque-Bera (JB):              719.382

Skew:                           1.212   Prob(JB):                    6.14e-157

Kurtosis:                       5.652   Cond. No.                         311.

==============================================================================

Notes:

[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.