Question: Pickles suggests that you use a wrapper method to perform feature selection. You decide to perform backward selection using Logistic Regression. First, you run the
Pickles suggests that you use a wrapper method to perform feature selection. You decide to perform backward selection using Logistic Regression. First, you run the full model, with all predictor variables as input. The output from the model is below. Based on the output, which of the following variables should you remove as a predictor variable?
Coefficients: Estimate Std. Error z value Pr(>|z|) (Intercept) -1.599e+00 1.412e-01 -11.323 < 2e-16 *** genderMale -4.197e-02 6.982e-02 -0.601 0.547788 SeniorCitizen1 4.956e-01 9.072e-02 5.463 4.68e-08 *** PartnerYes 8.208e-03 8.372e-02 0.098 0.921903 DependentsYes -3.158e-01 9.540e-02 -3.310 0.000933 *** tenure -6.765e-02 6.309e-03 -10.723 < 2e-16 *** PaperlessBillingYes 5.010e-01 7.902e-02 6.340 2.30e-10 *** MonthlyCharges 2.459e-02 1.993e-03 12.341 < 2e-16 *** TotalCharges 1.703e-04 7.043e-05 2.417 0.015631 * --- Signif. codes: 0 *** 0.001 ** 0.01 * 0.05 . 0.1 1
(Dispersion parameter for binomial family taken to be 1)
Group of answer choices
MonthlyCharges
Partner
TotalCharges
tenure
Step by Step Solution
There are 3 Steps involved in it
Get step-by-step solutions from verified subject matter experts
