#fitting multiple LR to train set from sklearn.linear_model import LinearRegression regressor=LinearRegression() regressor.fit(X_train,y_train) #predicting the test set y_pred=regressor.predict(X_test) #build the optimum model using bakcward elimination import statsmodels.formula.api as sm """adding one column for b0 which is constant in front as x0=1 """ X= np.append(arr= np.ones((50,1)).astype(int), values=X, axis=1) """step 2 and 3 of BE Model""" X_opt= X[:, [0,1,2,3,4,5]] regressor.OLS= sm.OLS(endog= y, exog= X_opt).fit() regressor.OLS.summary() """step 4 and 5 (remove the index 2 with highest p)""" X_opt= X[:, [0,1,3,4,5]] regressor.OLS= sm.OLS(endog= y, exog= X_opt).fit() regressor.OLS.summary() """repeating until we get FIN""" X_opt= X[:, [0,3,4,5]] regressor.OLS= sm.OLS(endog= y, exog= X_opt).fit() regressor.OLS.summary() X_opt= X[:, [0,3,5]] regressor.OLS= sm.OLS(endog= y, exog= X_opt).fit() regressor.OLS.summary() X_opt= X[:, [0,3]] regressor.OLS= sm.OLS(endog= y, exog= X_opt).fit() regressor.OLS.summary()
