Linear Regression 13 | Python for MLR Model Diagnosis — Part 3

Linear Regression 13 | Python for MLR Model Diagnosis — Part 3

1. Import Packages

import copy
from math import *

import pandas as pd
import numpy as np
import scipy
from scipy import stats
from scipy.stats import kstest
from scipy.stats import boxcox
import scipy.linalg as linalg

from sklearn import linear_model
import statsmodels.api as sm
import statsmodels.formula.api as smf
from statsmodels.stats.diagnostic import het_breuschpagan
from patsy import dmatrices
from statsmodels.stats.outliers_influence import variance_inflation_factor

import matplotlib.pyplot as plt
import seaborn as sns
%config InlineBackend.figure_format='retina'

2. Subset Selection

An example instance is,

SubsetSelection(df_brand, 'BrandLiking')

The output is,

3. Forward Stepwise Model Selection (Automatically Select)

An example instance is,

ForwardSelection(df_icecream, 'cons', 'R_sq_adj')

The plot of the model selection is,

The output is,

'cons ~ temp + income + price'

Note that different methods can give different results,

ForwardSelection(df_icecream, 'cons', 'aic')

The plot of the model selection is,

The output is,

'cons ~ income + price + time + temp'