Here is an example outline of my dataset:

  • I have 20 shops that are open for a period of time, but the duration is not the same (i.e. some locations are open for 4 years, others are open for 2 months, others again for 7 years and 2 months, etc)
  • In each location I have one of three possible sellers: "top", "adequate", "borderline".
  • These sellers are present at the different shops at different times (i.e. in one location there might be a "top" seller for 3 months, then an "adequate" seller for 1 year, then a "top" seller for 2 months, etc).

I am unsure how to best represent the risks of getting a borderline seller over time.

At first I thought a proportional stacked area chart would be suitable (e.g. the type shown here), but as the locations are open for different periods over time (e.g. there might only be two locations open for more than 3 years), this would introduce confounding and misrepresent the comparable risk.

I've been told hazard ratios or competing risks might be appropriate here?

Any help much appreciated.


Your Answer

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

Browse other questions tagged or ask your own question.