论文信息 - On Bartlett’s Formula for Non‐linear Processes

On Bartlett’s Formula for Non‐linear Processes

A Bartlett-type formula is proposed for the asymptotic distribution of the sample autocorrelations ofnonlinear processes. The asymptotic covariances between sample autocorrelations are expressed as thesum of two terms. The first term corresponds to the standard Bartlett’s formula for linear processes,involving only the autocorrelation function of the observed process. The second term, which is specificto nonlinear processes, involves the autocorrelation function of the observed process, the kurtosis of thelinear innovation process and the autocorrelation function of its square. This formula is obtained under asymmetry assumption on the linear innovation process. An application to GARCH models is proposed.

Christian Francq | Alain Berlinet | A. Berlinet | C. Francq

[1] R. Engle. Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation , 1982 .

[2] C. Granger,et al. An introduction to bilinear time series models , 1979 .

[3] T. Ozaki,et al. Modelling nonlinear random vibrations using an amplitude-dependent autoregressive time series model , 1981 .

[4] Joseph P. Romano,et al. Inference for Autocorrelations under Weak Assumptions , 1996 .

[5] H. Tong. Non-linear time series. A dynamical system approach , 1990 .

[6] M. Bartlett. On the Theoretical Specification and Sampling Properties of Autocorrelated Time‐Series , 1946 .

[7] Richard A. Davis,et al. Time Series: Theory and Methods , 2013 .

[8] T. Bollerslev,et al. Generalized autoregressive conditional heteroskedasticity , 1986 .