摘要
A robustified residual autocorrelation is defined based on L1-regression. Under very general conditions, the asymptotic distribution of the robust residual autocorrelation is obtained. A robustified portmanteau statistic is then constructed which can be used in checking the goodness-of-fit of AR(p) models when using L1-norm fitting. Empirical results show that L1-norm estimators and the proposed portmanteau statistic are robust against outliers, error distributions, and accuracy for a given finite sample.
A robustified residual autocorrelation is defined based on L<sub>1</sub>-regression. Under very general conditions,the asymptotic distribution of the robust residual autocorrelation is obtained. A robustified portmanteau statistic is then constructed which can be used in checking the goodness-of-fit of AR（p） models when using L<sub>1</sub>-norm fitting. Empirical results show that L<sub>1</sub>-norm estimators and the proposed portmanteau statistic are robust against outliers, error distributions, and accuracy for a given finite sample.
基金
Project supported by the Foundation of State Educational Commission
a research grant from the Doctoral Program Foundation of China (#97000139)
