Nonlinear Filters Based on Density Approximation
In the last chapter, we have analyzed which kinds of approximation have to be used in the nonlinear filters that utilize the Taylor series expansion. In addition to linear approximation of the nonlinear measurement and transition equations, we need to approximate the nonnormal error terms (or residuals) as the normal ones and the correlated error terms (or residuals) as the uncorrelated ones. These approximations are very strong because it is known that Kaiman filter models based on normality assumption are non-robust (see Meinhold and Singpurwalla (1989)). Therefore, density approximation, rather than function approximation, has to be investigated in order to obtain less biased filtering estimates of the state-variables.
KeywordsLikelihood Function Density Approximation Gibbs Sampler Extended Kalman Filter Taylor Series Expansion
Unable to display preview. Download preview PDF.