|
Abstract:
|
The talk will open with a review and unification of bias reduction methods in general, and for measurement error models in particular. Then I will explain how the common principles of bias reduction can be used to address standard and not-so-standard measurement error regression models; and also to address problems that are not usually regarded as measurement error problems per se, but that are, in a broad sense, generalized measurement error models (GMEM). Applications to measurement error regression modeling, variable selection (in non-measurement error regression models), and estimating degrees of freedom will be used to motivate, explain and illustrate the bias reduction strategy.
|
