Robustness of estimators in a finitely additive white noise model
Hucke Hans-peter · Stochastics · 1987
A finetely additive model is used for the non-linear estimation of a random signal in the presence of "white" Gaussian noise. The continuous dependence of the resulting estimators on the observed sample path and a priori parameters is investigated. It is shown that the estimators are robust in the sense that they are Lipschitz-continuous functions. Applications to the non-linear filtering problem and discrete approximations are also given