Table 1 Model parameters for DGP;, i € {1, 2, 3, 4, 5, 6}. Table 1
Related Figures (1)
Empirical levels (in percentage) of the Portmanteau test statistics @y and @;,, defined by (11) and (12), respectively, for the DGPj,i € {1, 2, 3, 4, 5, 6}. than @y, particularly as M increases. We concentrate the rest of our discussion on @;, only. Generally, the x? distribution provided a satisfactory approximation for all lags, at both significance levels. The results for the models without and with parameter constraints were very comparable. In general, the rejection rates of the test statistic Qj, at the 5% and 10% nominal levels lie within the 5% significant limits, or they are reasonably close to these intervals. It should be noted that some overrejection has been observed under multiplicative models, particularly for large values of M. This suggests that large sample sizes s may be needed for complex time series models with a 1 multiplicative structure. as . Se ee en se oe: og oe a _ oe ace cee AE pe eect eye cece gga cag H Table 2