Premium
A Study of Autoregressive and Window Spectral Estimation
Author(s) -
Beamish N.,
Priestley M. B.
Publication year - 1981
Publication title -
journal of the royal statistical society: series c (applied statistics)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.205
H-Index - 72
eISSN - 1467-9876
pISSN - 0035-9254
DOI - 10.2307/2346656
Subject(s) - autoregressive model , window (computing) , estimation , statistics , star model , spectral density estimation , mathematics , econometrics , computer science , autoregressive integrated moving average , time series , engineering , fourier transform , mathematical analysis , operating system , systems engineering
S ummary This paper describes the results of a simulation study aimed at comparing the relative merits of ar (autoregressive) and “Window” spectral estimation for stationary time series. Seven models are considered, namely ar (2), ar (4), ar (5), arma (2,2), ma (1) (two cases) and a process with a “mixed” spectrum. The paper also includes some discussion of two different methods of estimating the coefficients of ar models (the Burg method and the Yule–Walker approach), and of the performance of various order determination criteria, such as FPE, AIC and CAT.