What are good tests to check if the data is generated by multivariate jointly gaussian distribution.

I know $\chi^2$ test is commonly used as a goodness of fit metric but I did not understand how to use it for multidimensional random vector case. Most of the examples I saw were for univariate cases.

An example would help a lot.

Note: the covariance matrix isn't diagonal. No assumption of independent variables.


1 Answer 1


You can use the Mardia's Test.

The test has R package : https://www.rdocumentation.org/packages/MVN/versions/4.0/topics/mardiaTest (https://www.youtube.com/watch?v=pBJNXhl4cxQ&feature=youtu.be)


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.