Estimates models that extend the standard GLM to take misclassification into account. The models require side information from a secondary data set on the misclassification process, i.e. some sort of misclassification probabilities conditional on some common covariates. A detailed description of the algorithm can be found in Dlugosz, Mammen and Wilke (2015) https://www.zew.de/publikationen/generalised-partially-linear-regression-with-misclassified-data-and-an-application-to-labour-market-transitions.