2
$\begingroup$

I'm working with a data set in stata that contains information on how much each customer spent on service a, b, c, and d. But most customers don't have every service and this has resulted in a lot of zeros being counted and this might affect the finding of the analysis. Is there a statistical solution to this? And if yes, how do I use it to find which services are most profitable

$\endgroup$
3
  • $\begingroup$ The answer to this depends a lot on what you are trying to do, i.e. what kind of analysis are you interested in? $\endgroup$ – ekvall Apr 1 '16 at 13:22
  • $\begingroup$ @Student001 the doc file contains all the details $\endgroup$ – user110647 Apr 1 '16 at 13:25
  • $\begingroup$ Thanks @user110647, but please include all relevant information in the question itself. Also, you should probably use the self-study tag and read its wiki stats.stackexchange.com/tags/self-study/info. $\endgroup$ – ekvall Apr 1 '16 at 13:26
0
$\begingroup$

You don't have to use the cases where customers spent nothing on a particular service. Depending on what program you use, you can filter to only include cases where customers spent more than 0 £/$/etc on a service and run your analysis on the filtered data set.

If you want to account for the differences in use between services, you could calculate average spend per (paying) customer for each service and then weight it, for example by percentage of customers using this service. But if you wanted to examine how profitable a service is, you will also need to include costs and you will have to consider which part of these costs is fixed and which will change dependent on the number of customers using it.

$\endgroup$
2
  • $\begingroup$ Alright that worked fine. Is there anyway I can identify customers who spent much more or much less on each service. I'm using stata $\endgroup$ – user110647 Apr 1 '16 at 14:19
  • $\begingroup$ I don't use Stata so I can't comment on the actual functions but it sounds as if you're looking for outliers. Perhaps if you calculated your mean and your standard deviation of cost for each service, and then looked for customers whose spending is more than a standard deviation in either direction (i.e., above or below the mean)? $\endgroup$ – Serafina Apr 1 '16 at 15:09

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.