I have a dataset that contains factors corresponding income ranges of sampled persons, like people with factor 1 earn between 10,000 to 20,000, 2 between 20,000 to 30,000 . I could just make dummies for those ranges. But the factor variable for income is right censored. The norm is adding an extra dummy for censored/uncensored category, but this would lead to bias, according to some published papers. What is the best way to approach this problem? The dependent variable is discrete choice.

  • $\begingroup$ btw, my goal is inference... $\endgroup$ – cf121 Jan 3 at 2:58
  • $\begingroup$ Chi-square test should fit your needs if you bin your variables into factors. $\endgroup$ – Arun Jose Jan 3 at 8:27

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.