3. Consider seasonal aspects. An examination of the ACF and PACF at the seasonal lags can help identify AR and MA models for the seasonal aspects of the data, but the indications are by no means as easy to ¯nd as in the case of the non-seasonal aspects. For quarterly data, the forecaster should try to see the pattern of r4, r8, r12, r16, and so on, in the ACF and PACF.

For monthly data, it is seldom possible to examine very many autocorrelations for lags in multiples of 12. Thus r12, r24, and possibly r36 may be available|but these are all that can be used.