作者: M. Arató , G. Pap , K. Varga
DOI: 10.1016/S0898-1221(01)00315-7
关键词: Mathematics 、 Multivariate statistics 、 Component (thermodynamics) 、 Estimator 、 Applied mathematics 、 Autoregressive model 、 Statistics 、 Characteristic equation 、 Maximum likelihood 、 Wiener process
摘要: In this paper, we show that for autoregressive processes the estimators of mean are consistent if component process is ‘periodical’, and it not case a damping one. one-dimensional AR(1) case, cannot be estimated well. complex AR(1), where behaves periodically, can For an AR(2) process, well roots characteristic equation complex.