I want to ask about seasonal ARIMA (SARIMA) in R function how to determine that time series data has affected or influenced by seasonal pattern

Thank you very much

  • $\begingroup$ Look at the help file of function auto.arima of "forecast" package in R. There you will find a couple of tests for seasonality. $\endgroup$ – Richard Hardy Dec 2 '16 at 8:59

Pradita, You will need to go beyond auto.arima to find seasonality. If only one period has seasonality(ie June) then auto.arima won't find it. You need to consider more than just seasonal differencing, 11 monthly dummies or an AR12 and consider the possibility for the need of only 1 monthly dummy needing to exist(or maybe 2?). If there is a change in the seasonality(similar concept---Monday is low and then half way into the series Friday's become high) then auto.arima won't find it. If you have an example data set, post it here.


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.