Stable Autoregressive Models and Signal Estimation |
| |
Authors: | N. Balakrishna G. Hareesh |
| |
Affiliation: | 1. Department of Statistics , Cochin University of Science and Technology , Cochin , India balajicusat@yahoo.com;3. Institute for Systems Studies and Analyses , Metcalfe House , Delhi , India |
| |
Abstract: | This article studies the problem of model identification and estimation for stable autoregressive process observed in a symmetric stable noise environment. A new tool called partial auto-covariation function is introduced to identify the stable autoregressive signals. The signal and noise parameters are estimated using a modified version of Generalized Yule Walker type method and the method of moments. The proposed methods are illustrated through data simulated from autoregressive signals with symmetric stable innovations. The new technique is applied to analyze the time series of sea surface temperature anomaly and compared with its Gaussian counterpart. |
| |
Keywords: | Extended Yule Walker estimation Model identification Partial auto-covariation Signal estimation Stable autoregressive model |
|