Scholion.ai

Asymptotic Properties of the Wiener–Kolmogorov Predictor. Ii

R. J. Bhansali

Journal of the Royal Statistical Society: Series B (Statistical Methodology) · 1977 · Volume 39 · Issue 1

DOI: 10.1111/j.2517-6161.1977.tb01606.x

Abstract

The asymptotic properties of the autoregressive estimatesa(T)(u) obtained using the fles method (Bhansali, 1974) suggest that, in practice, a decision on truncating thea(T)(u) atu=p, say, will need to be taken. This truncated sequence of the auto-regressive estimates is used to form alternative estimates of the linear predictor, the autoregressive spectral density function and the innovation variance. Some of the asymptotic properties of these estimates are derived for the case when the observed time series is a sample of an infinite, as well as a finite, order autoregressive process. A comparison of these estimates with the corresponding estimates obtained using the time domain autoregressive coefficients is carried out.

Keywords

autoregressive process, factorization of spectral density function, exponential model, linear least square prediction, time series

Enable JavaScript for citation tools and additional paper interactions.