An Improved Estimator for a Gaussian AR(1) Process with an Unknown Drift and Additive Outliers

Authors

  • Wararit Panichkitkosolkul Department of Mathematics and Statistics, Faculty of Science and Technology, Thammasat University, Phathum Thani, 12121, Thailand.

Keywords:

additive outliers, AR(1) model, parameter estimation, recursive median

Abstract

This paper presents a new estimator for a Gaussian AR(1) process with an unknown drift and additive outliers. We apply the improved recursive median adjustment to the weighted symmetric estimator of Park and Fuller [1]. We consider the following estimators: the weighted symmetric estimator (\inline \hat{\rho}_{W}), the recursive mean adjusted weighted symmetric estimator (\inline \hat{\rho}_{R-W}) proposed by Niwitpong [2], the recursive median adjusted weighted symmetric estimator (\inline \hat{\rho}_{RMD-W}) proposed by Panichkitkosolkul [3] and the improved recursive median adjusted weighted symmetric estimator (\inline \hat{\rho}_{IRMD-W}). Using Monte Carlo simulations, we compare the mean square error (MSE) of estimators. Simulation results have shown that the proposed estimator, \inline \hat{\rho}_{IRMD-W}, provides a MSE lower than those of \inline \hat{\rho}_{W}, \inline \hat{\rho}_{R-W} and \inline \hat{\rho}_{RMD-W} for almost all situations.

Downloads

How to Cite

Panichkitkosolkul, W. (2015). An Improved Estimator for a Gaussian AR(1) Process with an Unknown Drift and Additive Outliers. Thailand Statistician, 8(1), 1–15. retrieved from https://ph02.tci-thaijo.org/index.php/thaistat/article/view/34293

Issue

Vol. 8 No. 1 (2010): JANUARY

Section

Articles