The lag 1 autocorrelation (r1) is a pattern indicator|it refers lag 1 ACF to the pattern of

² The lag 1 autocorrelation (r1) is a pattern indicator|it refers lag 1 ACF to the pattern of the errors. If the pattern is random, r1 will be around 0. If there are runs of positive errors alternating with runs of negative errors, then r1 is much greater than 0

(approaching an upper limit of 1). If there are rapid oscillations in errors (from positive to negative), then r1 is much less than 0 (approaching a lower limit of ¡1). Note that a value near 0 is not necessarily \best." For example, row 4 has r1 smaller than the minimum MSE model in row 8. But the row 4 model

(Pegels' A-2) does not include a trend component and is not a good model for these data.