The goal of this assignment is to familiarize you with running simple logit regressions in Stata. You will learn how to transform raw survey data into an appropriate format and how to interpret the coefficients from the
logit regression. In this hypothetical situation, you work for P&G and have been assigned a project where you need to quantitatively assess the willingness-to-pay (WTP) for various brands of detergent, WTP for extra detergent size, as well as the impact of income on detergent purchase. You have access to the survey data from 59 subjects, each answering 8 choice questions. The file containing the data on subjects’ income and their choices is called HW2 Excel Database.xls. The file with the survey questions that the subjects answered is called HW2 Survey.xls.
First, save this data file on your computer, open it in Excel and delete the first x observations (columns), where x is the last digit of your student number. So, if your student number ends with 5, you should delete the first 5 columns. My last digit number is 7.
The data does not need to be “cleaned”, but you do need to merge together the data file with the survey file to get the data into the appropriate format for running clogit command in stata. In class, I described the format that you need to have, your Stata sessions (see below) covered it, and, finally, I put a ‘clogit.pdf’ document on Ted that also describes the process (in part 4)
Afterwards, import your data into Stata, run the appropriate clogit command, first without and then with the income variable. Your write-up should fit on one page. It should contain one table that includes the
information from both regressions you ran (check textbooks and academic articles for examples of presenting multiple regressions in the same table). You should also include a short write-up of your findings: the interpretation of coefficients, discussion of their significance and discussion of the effect of including income into your model.
Last point, the goal of this assignment is to teach you how to troubleshoot such issues on your own, so try to do as much as you can by yourself. Use your class notes, the document I posted on Ted and online resources.
P.S. Please normalize the coefficients (both the brand dummy and the income coefficient) on Surf to zero.
P.P.S. Don’t forget to normalize the prices by the size of the detergent.