References & Citations
Mathematics > Statistics Theory
Title: Improved multivariate portmanteau test
(Submitted on 2 Nov 2016)
Abstract: A new portmanteau diagnostic test for vector autoregressive moving average (VARMA) models that is based on the determinant of the standardized multivariate residual autocorrelations is derived. The new test statistic may be considered an extension of the univariate portmanteau test statistic suggested by Pena and Rodriguez (2002, A Powerful Portmanteau Test of Lack of Test for Time Series, Journal of American Statistical Association) The asymptotic distribution of the test statistic is derived as well as a chi-square approximation. However, the Monte-Carlo test is recommended unless the series is very long. Extensive simulation experiments demonstrate the usefulness of this test as well as its improved power performance compared to widely used previous multivariate portmanteau diagnostic check. Two illustrative applications are given.
Link back to: arXiv, form interface, contact.