We propose a new sequential procedure for estimating multivariate distributions in cases when conventional maximum likelihood has too many parameters and is therefore inaccurate or non-operational. The procedure constructs a multivariate distribution and its pseudo-likelihood sequentially, in...

We numerically evaluate asymptotic variances and biases of various method-of-moments estimators in the Hansen-Singleton model calibrated to real data. Inspection of resulting figures leads to a conclusion that applied researchers do not always form instrument sets judiciously.

In models with many instruments, the asymptotic variance of the LIML estimator contains four components. Apart from the traditional variance, one term is due to instrument numerosity, and the last two appear if the model errors are non-normal. For a stylized instrumental variables model, we...

Applied researchers often use tests based on contingency tables, especially in preliminary data analysis and diagnostic testing. We show that many such tests may be alternatively implemented by testing for coefficient restrictions in linear regression systems.