So far, runs test or bds test are doing OK with i.i.d data but not so for independent but not identically distributed data set based on my practical experiences with them. Are there any other tests that have good properties at test these non-i.i.d data stream?

  • 1
    $\begingroup$ This question is not self-contained enough, in its present form, to be answerable. I think it would be helpful to supply a little more explanation and context. $\endgroup$ – cardinal Jan 20 '13 at 23:24
  • $\begingroup$ @cardinal Well, I am mostly interested in finding a test to test time-series data against a zero-mean and independantly distrubted stochastic processes (for instance, a random variable at time t: Xt~N(0, f(t)), where f(t) is some non-constant function), practical experiences shows BDS and runs tests works poorly in such cases, as the authors of these methods admitted. $\endgroup$ – user55647 Jan 20 '13 at 23:42

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.