A Note on the Strong Mixing Property for a Random Coefficient Autoregressive Process

  • Lee, Sang-Yeol (Department of Statistics, Sookmyung Women's University, Yongsan-ku, Seoul 140-742)
  • 발행 : 1995.06.01

초록

In this article we show that a class of random coefficient autoregressive processes including the NEAR (New exponential autoregressive) process has the strong mixing property in the sense of Rosenblatt with mixing order decaying to zero. The result can be used to construct model free prediction interval for the future observation in the NEAR processes.

키워드

참고문헌

  1. Journal of Time Series Analysis v.12 Estimation of the parameters of an EAR(p) process Billard,L.;Mohamed,F.Y.
  2. Annals of Statistics v.10 Nonparametric interval and point prediction using data trimmed by Grubb-type outlier rule Butler,R.W.
  3. Annals of Statistics v.15 Model free one step ahead prediction intervals:Asymptotic theory and sample simulations Cho,S.;Miller,R.B.
  4. Theory of Probability and its Applications v.22 On the strong mixing property for linear sequences Gorodetskii,V.V.
  5. Journal of the Royal Statistical Society Series, B v.47 Modelling and residual analysis of nonlinear autoregressive time series in exponential variables Lawrance,A.J.;Lewis,P.A.W.
  6. Journal of Econometrics v.10 Experience with using the Box-Cox transformation when forecasting economic time series Nelson,H.L.;Granger,C.W.J.
  7. Journal of the Royal Statistical Society Series, B v.48 Maximum likelihood estimation for the NEAR(2) model Smith,R.L.
  8. Journal of Korean Statistical Society v.17 A Laplacian autoregressive time series model Son,Y.S.;Cho,S.