Raphael was interested in understanding the factors related to churn for a telco provider. The available database had only a limited set of variables, so he was limited to looking at long distance, toll free, equipment, calling card, and wireless usage for the past month (all measured as scalar or metric variables) and whether the customer had multiple lines, voice mail, paging, internet, caller ID, call waiting, call forwarding, 3-way calling, and electronic billing services (measured as dummy yes/no variables). He ran a logistic regression with “churn last month” as the dependent variable and got the output displayed below.

 

i) (5 points) Why did Raphael decide to oversample customers who had churned last month to run the model?
ii) (5 points) Which variables appear to be significantly related to churn (if 0=no and 1=yes) at the .05 significance level?
iii) (4 points) Which factors appear to decrease the probability that a customer churned?
iv) (6 points) Outline the steps that Raphael might take to develop a churn management program using this model.

Transcribed Image Text:

Step 1 Long distance last month Toll free last month Equipment last month Calling card last month Wireless last month Multiple lines Voice mail Paging service Internet Caller ID Call waiting Call forwarding 3-way calling Variables in the Equation B S.E. Electronic billing Constant Wald 47.667 .062 6.619 848 067 .249 5.381 080 6.682 413 2.278 .778 2.897 2.295 1.929 df 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Sig. -126 .018 .000 -.002 .007 .803 016 .006 010 -.007 008 357 -.002 .007 .795 -.095 .191 618 529 228 020 .072 .257 .778 525 .203 010 153 237 521 -.358 237 .131 207 234 .378 -.388 228 089 .305 .202 .130 -.275 198 165 a. Variable(s) entered on step 1: Long distance last month, Toll free last month, Equipment last month, Calling card last month, Wireless last month, Multiple lines, Voice mail, Paging service, Internet, Caller ID, Call waiting, Call forwarding, 3-way calling, Electronic billing. Exp(B) 882 998 1.016 .993 998 909 1.697 1.075 1.690 1.165 .699 1.230 .679 1.357 759 Step 1 Long distance last month Toll free last month Equipment last month Calling card last month Wireless last month Multiple lines Voice mail Paging service Internet Caller ID Call waiting Call forwarding 3-way calling Variables in the Equation B S.E. Electronic billing Constant Wald 47.667 .062 6.619 848 067 .249 5.381 080 6.682 413 2.278 .778 2.897 2.295 1.929 df 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Sig. -126 .018 .000 -.002 .007 .803 016 .006 010 -.007 008 357 -.002 .007 .795 -.095 .191 618 529 228 020 .072 .257 .778 525 .203 010 153 237 521 -.358 237 .131 207 234 .378 -.388 228 089 .305 .202 .130 -.275 198 165 a. Variable(s) entered on step 1: Long distance last month, Toll free last month, Equipment last month, Calling card last month, Wireless last month, Multiple lines, Voice mail, Paging service, Internet, Caller ID, Call waiting, Call forwarding, 3-way calling, Electronic billing. Exp(B) 882 998 1.016 .993 998 909 1.697 1.075 1.690 1.165 .699 1.230 .679 1.357 759

Answer rating: 100% (QA)

1 The steady loss rate isnt near half At the end of the day the quantity of clients beat somewhat re... View the full answer