DIAGNOSTIC CHECKING ARMA TIME SERIES MODELS USING SQUARED-RESIDUAL AUTOCORRELATIONS |
| |
Authors: | A I McLeod W K Li |
| |
Affiliation: | Department of Statistical and Actuarial Sciences, The University of Western Ontario;Department of Statistics, University of Hong Kong |
| |
Abstract: | Abstract. Squared-residual autocorrelations have been found useful in detecting nonlinear types of statistical dependence in the residuals of fitted autoregressive-moving average (ARMA) models (Granger and Andersen, 1978; Miller, 1979). In this note it is shown that the normalized squared-residual autocorrelations are asymptotically unit multivariate normal. The results of a simulation experiment confirming the small-sample validity of the proposed tests is reported. |
| |
Keywords: | ARMA time series diagnostic checking nonlinear time series portmanteau test testing for statistical independence |
|