ABSTRACT
AR:-JA model may then be fitted to the differenced series as described in Sections 4.2 4.4. It is often found that first-order differencing of nonseasonal data is adequate - sec Section 3.4.6 and Example 14.2 - although second-order differencing is occasionally required. The resulting model for the undifferenced series is the fitted ARII'viA model. For seasonal data, se<ct..sonal differencing may also be required (see Section 4.6 and Example 14.3).
