Previous |  Up |  Next


randomcoefficient autoregression; heteroskedasticity; wild bootstrap
In the paper, a heteroskedastic autoregressive process of the first order is considered where the autoregressive parameter is random and errors are allowed to be non-identically distributed. Wild bootstrap procedure to approximate the distribution of the least-squares estimator of the mean of the random parameter is proposed as an alternative to the approximation based on asymptotic normality, and consistency of this procedure is established.
[1] Anděl J.: Autoregressive series with random parameters. Math. Operationsforch. Statist. 7 (1976), 735–741 DOI 10.1080/02331887608801334 | MR 0428649
[2] Davidson J.: Stochastic Limit Theory. Oxford University Press, New York 1994 MR 1430804 | Zbl 0904.60002
[3] Janečková H.: Time Series with Changing Parameters. Ph.D. Thesis, Charles University, Prague 2002
[4] Janečková H.: Some generalizations in a heteroscedastic RCA(1) model. Acta Univ. Carolin. – Math. Phys. 43 (2002), 21–47 MR 1959860 | Zbl 1186.62107
[5] Jürgens U.: The estimation of a random coefficient AR(1) process under moment conditions. Statist. Hefte 16 (1985), 237–249 DOI 10.1007/BF02932535 | MR 0861799 | Zbl 0573.62086
[6] Kreiss J. P.: Asymptotical properties of residual bootstrap for autoregression. Preprint, TU Braunschweig 1997
[7] Liu R. Y.: Bootstrap procedures under some non-i. i.d. models. Ann. Statist. 16 (1988), 1696–1708 MR 0964947 | Zbl 0655.62031
[8] Nicolls D. F., Quinn B. G.: Random Coefficient Autoregressive Models: An Introduction (Lecture Notes in Statistics 11). Springer–Verlag, New York 1982 MR 0671255
[9] Prášková Z.: Bootstrapping in nonstationary autoregression. Kybernetika 38 (2002), 389–404 MR 1937136
Partner of
EuDML logo