scientific article; zbMATH DE number 3898069

From MaRDI portal
Publication:3677010





A method to find the best linear prediction of q-th order autoregressive time series having incomplete information is derived. If \(\{Y_ t\}\) is a q-th order autoregressive time series which can be represented as a moving average process, \(E(Y_ t)=0\), the autocovariation function of the series and the realization of the time series \(\{Y_ tX_ t\}\) are known, \(\{X_ t\}\) being a stationary time series independent of \(\{Y_ t\}\), under some natural conditions the best linear prediction of \(\{Y_ t\}\) is derived from a system of algebraic linear equations. The case of the first order autoregressive time series with zero mean is also exposed. The mean square error of prediction is given. The author states the existence of an analogous result for the prediction of a finite moving average process.




Cited in
(23)








This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3677010)