As discussed before, removing seasonal component is in general a good practice for pre-processing data in statistical modeling.

My particular question comes from the fact that I'm working with three times series regarding the labor market in their raw form: unemployment rate (U/EAP), employment rate (E/WAP) and informality rate (I/E). Inspecting them visually with a seasonal plot (ggseasonplot), unemployment rate is clearly a very seasonal variable (low at the end of the year), but the other two do not exibit, at least visually, a seasonal component.

Then, should I deseasonalize time series that do not seem to have a very strong seasonal component?

Also, if I use the seasonal implementation of the X13-ARIMA-SEATS algorithm for seasonal adjustment, to a time series that has no seasonal component, will that reduce the quality or cause some undesired effect on the data?

If it's worth mentioning, my final aim is to get cyclical component to make business cycle analysis.



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.