Discussion of 'identification and estimation of nonlinear models using two samples with nonclassical measurement Errors' by Carroll et al.

Carroll et al. described a very general framework for problems involving measurement errors. The approach was based on the sieve likelihood principle where the dimension of the parameter space is allowed to grow with sample size. Thus the procedure includes many well-known estimators as special cases. In this discussion, a specific class of estimators based on B-splines will be described. It is observed that certain regularity conditions will be required to ensure the existence of the maximum likelihood estimators.