Data for Measurement Error in Generalized Linear Models

Key WOrds and Phrases: errors-in-variab7es; functiona7 mode7s; genera7ized 7inear mode7; /II-estimation; measurement error; structura7 mode7s. ABSTRACT This paper discussed a general strategy for reducing measurement-error-induced bias in statistical models. It is assu.ed that the measure.ent ·error is unbiased with a known variance although no other distributional assulIPtions on the measur..ent-error distribution are employed. Using a preliminary fit of the model to the observed data, a transformation of the variable measured with error is estimated. The transformation is constructed so that the estimates obtained by refitting the model to the "corrected" data yields estimates with smaller bias. Whereas the general strategy can be applied in a wide variety of settings, this paper focuses on the problem of covariate measurement error in generalized linear models. Two estimators are derived and their effectiveness at reducing bias is demonstrated in a Monte Carlo study.