Estimation for nonlinear stochastic differential equations by a local linearization method 1

This paper proposes a new local linearization method which approximates a nonlinear stochastic differential equation by a linear stochastic differential equation. Using this method, we can estimate parameters of the nonlinear stochastic differential equation from discrete observations by the maximum likelihood technique. We conduct the numerical experiments to evaluate the finite sample performance of identification of the new method, and compare it with the two known methods: the original local linearization method and the Euler methods. From the results of experiments, the new method shows much better performance than the other two methods particularly when the sampling interval is large

[1]  A. Harvey,et al.  The Estimation of Higher-Order Continuous Time Autoregressive Models , 1985, Econometric Theory.

[2]  G. S. Ladde,et al.  Euler-type approximation for systems of stochastic differential equations , 1989 .

[3]  A. Bergstrom.  Continuous Time Econometric Modelling. , 1992 .

[4]  Tohru Ozaki.  Identification of nonlinearities and non-Gaussianities in time series , 1992 .

[5]  Hirotugu Akaike,et al.  On entropy maximization principle , 1977 .

[6]  T. Ozaki.  A local linearization approach to nonlinear filtering , 1993 .

[7]  Peter V. Tryon,et al.  Continuous time series models for unequally spaced data applied to modeling atomic clocks , 1987 .

[8]  G. Box,et al.  Scientific Inference, Data Analysis and Robustness. , 1985 .

[9]  W. Rüemelin.  Numerical Treatment of Stochastic Differential Equations , 1982 .

[10]  Hirotugu Akaike,et al.  Statistical Inference and Measurement of Entropy , 1983 .

[11]  Tohru Ozaki,et al.  Statistical Identification of Storage Models with Application to Stochastic Hydrology , 1985 .

[12]  Karl Johan Åström,et al.  On the choice of sampling rates in parametric identification of time series , 1969, Inf. Sci..

[13]  Statistical Identification of Nonlinear Random Vibration Systems , 1989 .

[14]  Amir Dembo,et al.  On the parameters estimation of continuous-time ARMA processes from noisy observations , 1987 .

[15]  Hermann Singer,et al.  CONTINUOUS-TIME DYNAMICAL SYSTEMS WITH SAMPLED DATA, ERRORS OF MEASUREMENT AND UNOBSERVED COMPONENTS , 1993 .