Instructions
o Keep your answers concise. Use bullet points if applicable.
o Where applicable, provide your answer as well as the R code you used, outputs from the console, and any further calculations you conducted.
o For empirical analysis, the desired level of significance is 5%.
Questions
1. Why can’t we directly interpret the effects of the independent variables on the dependent variable in a logistic regression model? (2 points)
2. For the Customer Acquisition data:
a. Estimate a logistics regression model that include all the other variables as
IVs. Report the model output and interpret the effect of all IVs on the
probability of acquisition. (2 points)
Note: No need to calculate the odds ratio. Comment on direction and
statistical significance of the effect.
b. Compare this model with the model where only Acq_Expense is included
as IV and comment on the performance of this model. (2 points)
Note: Use pseudo R2 and AIC as criteria
Get Free Quote!
267 Experts Online