Kim Nolsøe, DTU

Estimating Functions with prior knowledge

Abstract

We will discuss a method for estimation of parameters which is
formulated in an estimating function setting. This method allows
the use of prior information. The main idea is to use prior
knowledge of the parameters, either specified as moments
restrictions or as a distribution, and use it in the construction
of an estimating function. It may be useful when the full Bayesian
analysis is difficult to carry out for computational reasons.
A brief introduction to estimation function theory will also be
presented.