Hypothesis Testing in Unidentified Models
Trevor Breusch · The Review of Economic Studies · 1986
An identified model is not necessary for statistical inference, but ambiguities can arise. This paper examines some simple examples and proposes a framework that distinguishes between the “refutation” and “confirmation” aspects of testing in an unidentified model. One particular problem is the interpretation given to overidentifying restrictions: a common view is that these are somehow not properly testable.