论文信息 - Hellinger distance estimation of nonlinear dynamical systems

Hellinger distance estimation of nonlinear dynamical systems

The present paper deals with the minimum Hellinger distance (MHD) parameter estimation of nonlinear dynamical systems with highly correlated residuals. In order to estimate the parameter of interest, we fit the residuals by an exponential autoregressive time-series model. Under some assumptions which ensure the stationarity, the existence of the moments of the stationary distribution and the strong mixing property of the fitted residuals, we establish the almost sure convergence and the asymptotic normality of the MHD estimates.

Ouagnina Hili | O. Hili

[1] P. Robinson. NONPARAMETRIC ESTIMATORS FOR TIME SERIES , 1983 .

[2] Sur l'estimation des modles autorgressifs d'ordre multiple de sries temporelles , 2001 .

[3] Georges Bastin,et al. A maximum likelihood parameter estimation method for nonlinear dynamical systems , 1999, Proceedings of the 38th IEEE Conference on Decision and Control (Cat. No.99CH36304).

[4] Peter Hall,et al. Cross-validation in density estimation , 1982 .

[5] Hili Ouagnina,et al. On the estimation of nonlinear time series models , 1995 .

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

[7] O. Hili. On the estimation of -ARCH models , 1999 .

[8] H. Kantz,et al. Nonlinear time series analysis , 1997 .

[9] M. Carbon. Inégalité de Bernstein pour les processus fortement mélangeants, non nécessairement stationnaires. Applications , 1983 .

[10] Georges Bastin,et al. An estimator of the inverse covariance matrix and its application to ML parameter estimation in dynamical systems , 2001, Autom..

[11] R. Beran. Minimum Hellinger distance estimates for parametric models , 1977 .