Financial advising
It often becomes tempting to add a large collection of variables to a regression, hoping to let regression "sort out" which variables are important rather than think about why each might matter. Beginning with the model shown in Table 4, add the ten variables that measure population in the ZIP Code of the business location to the model (Popu- lation of ZIP Code through Female Population).
(a) What happens when these are added to the model?
(b) How should these variables be added more sensibly? Does doing so produce a statistically significant improvement in the fit? Explain.