Title of article :
Parameter estimation of state space models for univariate observations
Author/Authors :
Costa، نويسنده , , Marco and Alpuim، نويسنده , , Teresa، نويسنده ,
Issue Information :
روزنامه با شماره پیاپی سال 2010
Abstract :
This paper contributes to the problem of estimation of state space model parameters by proposing estimators for the mean, the autoregressive parameters and the noise variances which, contrarily to maximum likelihood, may be calculated without assuming any specific distribution for the errors. The estimators suggested widen the scope of the application of the generalized method of moments to some heteroscedastic models, as in the case of state-space models with varying coefficients, and give sufficient conditions for their consistency. The paper includes a simulation study comparing the proposed estimators with maximum likelihood estimators. Finally, these methods are applied to the calibration of the meteorological radar and estimation of area rainfall.
Keywords :
Area rainfall estimates , Kalman filter , State space model , Parameters estimation
Journal title :
Journal of Statistical Planning and Inference
Journal title :
Journal of Statistical Planning and Inference