Large sample properties of parameter estimates for periodic ARMA models

From MaRDI portal





An ARMA model with coefficients which are periodic functions of the time (with known period) is considered. The authors demonstrate how to represent this model in a multivariate ARMA form. Weighted least squares estimators of the coefficients and the innovation variances are considered. Asymptotic normality of these estimators is demonstrated for causal invertible Gaussian models.




Cited in
(66)








This page was built for publication: Large sample properties of parameter estimates for periodic ARMA models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2784953)