Validation of state-space models from a single realization of non-Gaussian measurements

A methodology is presented for testing whether a dynamic model in linear state-space form accurately describes the system under consideration. Unlike existing procedures it is not necessary to assume that all of the random terms in the model are normally distributed. The methodology is based on a single realization of observations and is relatively easy to implement since it relies on a normalized Kalman filter state estimate. The testing procedure rests on an asymptotic distribution theory for the filter estimate.