0
$\begingroup$

What would be the proper way to determine statistically significant changes between time periods within a time series (between Yn and Yn+1)? I thought about taking the first difference and calculating the z value for the difference value.

$\endgroup$
  • $\begingroup$ Please consider editing to say what type of change you're trying to detect, the nature of your data, and any relevant assumptions. As a general hint for finding relevant literature, what you're trying to do is called "changepoint detection" $\endgroup$ – user20160 Jun 26 '18 at 4:12
  • $\begingroup$ Changepoint detection is what I was looking for. $\endgroup$ – Jeff Jun 26 '18 at 23:49
1
$\begingroup$

With provided details I think a self-extracting threshold model might server your purpose. If you are an R programming language user you can use tsDyn package. The setar function in the package can fit model to different regimes determined by significant change.

$\endgroup$

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.