I am going to decompose one of my time series. My goal is to get rid of the seasonal component. It looks like this:

Timeseries Log

The last picture is: log(next value/present value)

I have understood that if the seasonal variation is constant, you should use an additive model. And if the time series increases in magnitude, you should use an multiplicative model.

The graph is measuring income.

In my case it looks like there is a bit of both, so I can't figure out which one to use. Would anybody please help?




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.