I have a question about this question: Is this an appropriate method to test for seasonal effects in suicide count data?

and the first answer.

Why does using the $\cos$ and $\sin$ account for seasonality? I do not see why that is obvious, and the answer does not explain it.

  • 1
    $\begingroup$ Sin and cos are "wawy" functions, and seasonality can be seen as "wawes" $\endgroup$ – kjetil b halvorsen May 3 '17 at 11:44
  • 2
    $\begingroup$ This technique is often called "harmonic regression." If you perform an Internet search on it you will be able to find some more resources. Basically, the sin and the cos terms capture the oscillations of the dependent variable. Some extra computation using the regression coefficients of the cos and sin terms can further reveal other interesting numbers such as time to peak, height of peak, etc. $\endgroup$ – Penguin_Knight May 3 '17 at 13:39
  • $\begingroup$ I think this understates how much is said in the several answers and many comments to that thread, including their links. Trivial correction. @kjetilbhalvorsen means "waves". $\endgroup$ – Nick Cox May 3 '17 at 14:57

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.