A NOTE ON COVARIATE MEASUREMENT ERROR IN NONLINEAR MIXED EFFECTS MODELS

Little is known about the effects of measurement error in intra-indivi dual covariates on inference for the nonlinear mixed effects model. We investigate this issue for a controlled variable measurement error mo del, and find that a major consequence may be substantial bias in esti mates of parameters characterising intra-individual variation. Estimat ion of population parameters may also be affected. The dramatic effect of measurement error on estimation of intra-individual variance param eters has implications not only in the repeated measurement context, b ut-also in individual nonlinear regression models.