Say, I have two vector of percentages (in this particular case, percentages represent a proportion of certain cell types in blood samples). Both vectors sum up to 1.

For example, there are vectors A = [0.1, 0.15, 0.15, 0.20, 0.25, 0.15] and B = [0.05, 0.2, 0.15, 0.15, 0.2, 0.25], where A[1] and B[1] represent a proportion of cell type #1, etc.

What I want to do here is to find out whether the "composition" of cell types is similar between A and B. What would be the right statistical test to do this?


The data you have is two probability distributions over the 6 categories: they are all non-negative and they sum up to 1.

To tell if two distributions are the same over a set of categories (this is what you can call 'similar') you can use the Chi-squared test:


  • 1
    $\begingroup$ @sus_mlm: But please notice that you will need to know the size of the samples. $\endgroup$ – Pere Sep 19 '16 at 14:35

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.