The meaning of P value is probability which should be number between 0 (the event never occurs) and 1 (the event occurs always). significance testing for periodicity using Matlab gives a documentation on how to apply Fisher's statistics to a time series signal for periodicity testing. Some concepts are unclear to me and I shall be grateful for an intuitive answer:

Q1: In my example, I want to prove if the signal is periodic or not. So something that I want to prove to be true becomes the null hypothesis statement? Thus, would

 H0: the signal is periodic and H1: not periodic?

Q2: I got the value of pval for my signal as 2.3e-6 fmax (the maximum frequency) = 0.0078 andfisher_g = 0.1136`. How do I interpret this result? Is my signal periodic or not?

Q3: If the calculated pval is less than the threshold which usually is 0.05 then we reject the null hypothesis. But in the documentation it is the opposite and the threshold is 0.000001 which I do not know how it has come. I am confused! Please correct me where wrong.

  • 2
    $\begingroup$ I’m voting to close this question because this is a statistics question, not signal processing; stats.stackexchange.com might be a better place to ask. $\endgroup$ – Marcus Müller May 6 at 17:38

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.