I have sample-data for two multivariate normal distributions. From this sample-data, I can calculate each distribution’s parameters (means and standard deviations).

  1. How do I quantify the distance (or divergence) between these two multivariate normal distributions? I don’t mind whether I use the sample-data or the distribution parameters to make this quantification.
  2. Given the chosen test in (1), how do I calculate an accompanying p-value which can be used to reject the null hypothesis for the given pair of multivariate normal distributions?

While I can implement solutions in Python, I’m no stats expert, and would appreciate explanations formatted as such; for complex explanations, corresponding simplifications would be very useful.


1 Answer 1


I'm not sure whether I understand your question correctly, but from your description I think you might want to do a multivariate analysis of variance (MANOVA). It is used to compare sample means of multivariate distributions and returns p values as well.

Again, if I understand your question correctly, for an implementation you would need to assign a variable to each sample, which indicates whether the values belong to sample one or sample two. You could then use this variable as your independent variable and each of your sampled variables as a dependent variable.

  • $\begingroup$ Thanks, I will look into this. If there's anything I can do to clarify my questions, please let me know. $\endgroup$
    – Sam Huguet
    Jan 4, 2023 at 11:57
  • $\begingroup$ I think the question was great, I'm still a newbie as well and not used to answering questions. So the confusion comes from my side, not yours :) Hope my answer is helpful! $\endgroup$
    – David
    Jan 4, 2023 at 12:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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