Is there a way to test a hypothesis for a time series where $H_0 : \mu <$ threshold ?

For example, if I am monitoring an athlete's heart rate and there is a sudden spike in his heart rate from 100bpm to 200bpm, I want to test if the heart rate's mean is now higher than some average threshold it was previously at.

For example, differentiating between these two series:


Where my null hypothesis would be, following the spike I expect the heart to decrease below 150bpm on average unless given some contradicting evidence. (in reality my series are much longer but this toy example gets the point across).

I know there are mean reversion tests but I am not concerned with a return to some previous mean as I care if the heart rates are below a particular threshold that may or may not be related to the previous mean of the heart rates.

Are there methods aimed at this type of analysis?


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.