1
$\begingroup$

Let's say I am trying to compare the performance of a group of salesmen over the course of three years (not all have three full years of data). I have monthly data showing the number of customers, number of sales, and revenue for each. What I want to do is analyze these to see which salesmen are statistically outperforming or under performing the group.

The first thing I thought of was doing a t-test comparing each salesman to the average but I don't think that would work due to the time aspect. Is there a specific type of test/modeling technique I should look into?

$\endgroup$
1
$\begingroup$

To test for salesman within a group I would for each characteristic (# of cust, #of sales ,revenue) construct an ARIMA model that might include trends/level shifts,seasonal structure) while identifying potential anomalies. I would do this for each salesman in the group. I would then determine a common model by examining each of the models for similar structure. I would then estimate this model for each of the salesman separately and for the group as a whole. I would then construct an F test to test the hypothesis of a common set of parameters across salesman by pooling the error sos from each salesman vis-a-vis the error sos for the group composite series in the standard way.

If you have K groups with say Nk salesman in each group , then I would use the suggested procedure at that level rather than at the salesman within a group example I presented in the first paragraph..

$\endgroup$

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.