3
$\begingroup$

We know that red noise is the same as a first-order autoregressive (AR(1)) stationary Gaussian process with a positive correlation at unit lag. Does there exist some formal statistical test for red noise process?

One way I can think of is to firstly fit an AR(1) model to the process, then perform white noise test on the model residuals to see whether the residuals follow a white noise process. If yes, the original process can be viewed as a red noise process...

Welcome any ideas and comments.

$\endgroup$
  • $\begingroup$ Just interested, where can I read about why red noise is the same as AR(1) stationary Gaussian process with a positive correlation at unit lag? $\endgroup$ – Dmitry Laptev Oct 12 '12 at 14:42
  • $\begingroup$ @DmitryLaptev Please refer to: H. von Storch and F. W. Zwiers (1999), Statistical Analysis in Climate Research, Cambridge, England: Cambridge University Press. $\endgroup$ – Elaine Oct 13 '12 at 9:25

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.