This seminar will offer a  review the most common situations in which the regularity conditions that underlie classical likelihood-based parametric inference fail, focusing on the large-sample properties of the likelihood ratio statistic.  Three main classes of problems emerge: boundary problems, indeterminate parameter problems —- which include non-identifiable parameters and singular information matrices -- and change-point problems.  I'll emphasise analytical solutions, consider software implementations where available, and summarise how the key results are derived.

Coffee will be served at 10:30.