Variance and covariance computations for 2-D ARMA processes
The partial-fraction expansion procedure developed by Górecki and Popek is extended to the decomposition of a 2-D rational function whose denominators are symmetric with respect to the unit bi-circle. The extension is based on performing a one-variable bilinear transformation on the 2-D rational transfer function. This extension is used to develop an algorithm for evaluation of the complex integral \[ I= {1\over(2\pi i)^2} \oint_{| z_1|= 1} \oint_{| z_2|= 1} G(z_1,z_2) G(z^{-1}_1, z^{-1}_2) {dz_2 dz_1\over z_1z_2}, \] where \(i= \sqrt{-1}\) and \(G(z_1,z_2)= {B(z_1,z_2)\over A(z_1,z_2)}\) is a stable 2-D transfer function. The value of \(I\) is determined by the residue method and by finding the roots of a 1-D polynomial.
- An efficient and versatile algorithm for computing the covariance function of an ARMA process
- An evaluation of algorithms for computing the covariance function of a multivariable arma process
- Covariances between estimated autocorrelations of an ARMA process
- scientific article; zbMATH DE number 883617
- scientific article; zbMATH DE number 946687
- Computing the asymptotic covariance matrix of a vector of sample autocorrelations for ARMA processes
- On the estimation of variance for autoregressive and moving average processes (Corresp.)
- Characterization and estimation of two-dimensional ARMA models
- scientific article; zbMATH DE number 1515429
This page was built for publication: Variance and covariance computations for 2-D ARMA processes
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1289150)
