Regression Question for Measuring a Corporate Campaign

#1
Hi everyone,

I hope you are all having a great day. Currently, I am trying to measure the effect of an in-store campaign that deals with sampling items. I have 80 stores (40 control and 40 test). I have many interesting variables, such as average local income, number of competitors within X amount of miles, etc... This campaign will last for approximately one week and I'd like to measure the increase in revenue associated with this campaign. Any recommendations along with any assumptions I need to make sure hold?

Thanks!
 

noetsi

Fortran must die
#2
Well if you are running regression, or ANOVA, you have to meet all the standard Gauss Markov assumptions such as independence, no Multicolinearity, linearity etc. Any standard book on regression overs those.

I think the key would be to control for factors that might influence sales other than your test variable. Identifying and measuring these is going to be the most important task.