Abstract:
To estimate the unknown autoregression parameter in the case when noise has an infinite dispersion, the weighted estimate by the least-squares method is suggested. The limit distribution of the error of estimation is obtained. It is shown that the weighted estimate is asymptotically more exact in comparison with the common estimate by the least-squares method.
Presented by the member of Editorial Board:A. V. Nazin
Citation:
A. S. Markov, “Estimation of the autoregression parameter with infinite dispersion of noise”, Avtomat. i Telemekh., 2009, no. 1, 104–118; Autom. Remote Control, 70:1 (2009), 92–106