Kybernetika 32 no. 5, 491-499, 1996

Distribution of estimators and optimal experimental design in nonlinear regression

Andrej Pázman


In this survey we present in a condensed exposition good approximations of probability densities of LS estimators in nonlinear regression models. This includes also marginal densities or densities of scalar parametric functions. Further, the main ideas of two approaches how to use these densities for optimal experimental design are presented: the approach using the second order approximation technique, and the approach using optimality criteria in an integral form.


62K05, 62E17, 62J02