This paper gives a precise meaning to the robustness of risk-sensitive filters for problems in
which one is uncertain as to the exact value of the probability model. It is shown that risk-
sensitive estimators (including filters) enjoy an error bound which is the sum of two terms,
the first of which coincides with an upper bound on the error that one would obtain if one
knew exactly the underlying probability model, while the second term is a measure of the
distance between the true and design probability models. The paper includes a discussion …

… However, no robustness results for risksensitive filters have appeared to date in the
literature, to our knowledge. This paper gives a precise meaning to the robustness of risk-sensitive
filters. We assume that the true probabilistic model is fixed but unknown, and that the estimation
(or filtering) procedure makes use of a fixed nominal model. It is shown that …