Diagnostic checking ARMA time series models using squared‐residual autocorrelations
AI McLeod, WK Li - Journal of time series analysis, 1983 - Wiley Online Library
AI McLeod, WK Li
Journal of time series analysis, 1983•Wiley Online LibrarySquared‐residual autocorrelations have been found useful in detecting nonlinear types of
statistical dependence in the residuals of fitted autoregressive‐moving average (ARMA)
models (Granger and Andersen, 1978; Miller, 1979). In this note it is shown that the
normalized squared‐residual autocorrelations are asymptotically unit multivariate normal.
The results of a simulation experiment confirming the small‐sample validity of the proposed
tests is reported.
statistical dependence in the residuals of fitted autoregressive‐moving average (ARMA)
models (Granger and Andersen, 1978; Miller, 1979). In this note it is shown that the
normalized squared‐residual autocorrelations are asymptotically unit multivariate normal.
The results of a simulation experiment confirming the small‐sample validity of the proposed
tests is reported.
Abstract
Squared‐residual autocorrelations have been found useful in detecting nonlinear types of statistical dependence in the residuals of fitted autoregressive‐moving average (ARMA) models (Granger and Andersen, 1978; Miller, 1979). In this note it is shown that the normalized squared‐residual autocorrelations are asymptotically unit multivariate normal. The results of a simulation experiment confirming the small‐sample validity of the proposed tests is reported.
Wiley Online Library以上显示的是最相近的搜索结果。 查看全部搜索结果