I am trying to detect trends using non-parametric methods but I'm a little confused as to when you should apply the Seasonal Kendall test. Don't get me wrong I know you apply it when you have seasonality but at what point do you determine this effect insignificant? If I see seasonality in my data should I simply apply the Seasonal Kendall test instead of the non-seasonal Mann-Kendall test or ideally should I always be performing some sort of statistical test for seasonality like the Kruskal Wallis?

How does using the Mann_kendall test perform on data which is affected by seasonality?

  • $\begingroup$ Addint the tags time-series and seasonality may result in more views of your question and more potential answers. $\endgroup$ – javlacalle Feb 18 '15 at 19:00
  • $\begingroup$ Simon - If you have worked out some understanding of the issue, please feel free to post it here as an answer to your own question. There are others (myself included) who would be interested to read what you have come up with. $\endgroup$ – Adam Sep 3 '15 at 21:03

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.