COMPARATIVE POWER STUDIES FOR GOODNESS OF FIT TESTS OF TIME SERIES MODELS
- 1 May 1982
- journal article
- Published by Wiley in Journal of Time Series Analysis
- Vol. 3 (3) , 141-151
- https://doi.org/10.1111/j.1467-9892.1982.tb00336.x
Abstract
Power studies for three tests of fit, namely the classical test given by Whittle, the widely‐known test given by Ljung and Box, and the recent test given by Godolphin, are derived for the invertible time series models considered previously in an empirical study. The results demonstrate that it is possible to improve on the Ljung‐Box test by employing more empirically‐based alternatives. In particular, the test derived by Godolphin appears to be more powerful than the other tests. The results also suggest that the Whittle and Ljung‐Box tests are complementary, so that the combination of both may yield a sensitive test procedure.Keywords
This publication has 19 references indexed in Scilit:
- A method for testing the order of an autoregressive-moving average processBiometrika, 1980
- A Bayesian extension of the minimum AIC procedure of autoregressive model fittingBiometrika, 1979
- Some power studies of a portmanteau test of time series model specificationBiometrika, 1979
- On a Measure of Lack of Fit in Time Series ModelsBiometrika, 1978
- Estimating the Dimension of a ModelThe Annals of Statistics, 1978
- Some Properties of the Order of an Autoregressive Model Selected by a Generalization of Akaike's EPF CriterionBiometrika, 1977
- A new look at the statistical model identificationIEEE Transactions on Automatic Control, 1974
- Distribution of Residual Autocorrelations in Autoregressive-Integrated Moving Average Time Series ModelsJournal of the American Statistical Association, 1970
- Large-Sample Estimation of Parameters for Moving-Average ModelsBiometrika, 1961
- The Fitting of Time-Series ModelsRevue de l'Institut International de Statistique / Review of the International Statistical Institute, 1960