Flexible Parametric Measurement Error Models
Raymond J. Carroll, Kathryn Roeder and Larry Wasserman
Inferences in measurement error models can be sensitive to modeling
assumptions. Specifically, if the model is incorrect then the
estimates can be inconsistent. To reduce sensitivity to modeling
assumptions and yet still retain the efficiency of parametric
inference we propose to use flexible parametric models which can
accommodate departures from standard parametric models. We use
mixtures of normals for this purpose. We study two cases in detail: a
linear errors-in-variables model and a change-point Berkson model.
Keywords: Berkson model; change-point; Errors-in-variables;
Markov chain Monte Carlo; Normal mixture model
Here is the full postscript text for this
technical report. It is 292556 bytes long.