I am running ols regressions with firms from different countries and different industries. I want to control for both country and industry membership of each firm. Say I have 1,...,N countries and 1,...,M industries. How many dummy variables do I need to avoid the dummy trap? I am using Matlab, so I have to manually set up all regressors.

  • 1
    $\begingroup$ Is this a question from a course or textbook? If so, please add the [self-study] tag & read its wiki. $\endgroup$ – gung Apr 6 '16 at 18:05
  • $\begingroup$ No, its not a question from a course or textbook. However, I assume that I only need N-1 country dummies and M-1 industry dummies. Is this correct? $\endgroup$ – user111237 Apr 6 '16 at 18:10
  • 1
    $\begingroup$ Can you explain to the rest of us what the dummy trap is? $\endgroup$ – mdewey Apr 6 '16 at 20:12
  • $\begingroup$ What is your sample size? How many countries do you have and how many different industries? $\endgroup$ – Marquis de Carabas Apr 19 '16 at 4:44

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.