SEMIPARAMETRIC IDENTIFICATION AND FISHER INFORMATION

This paper provides a systematic approach to semiparametric identification that is based on statistical information as measure of its “quality.” Identification can be regular or irregular, depending whether the Fisher for parameter positive zero, respectively. I first characterize these cases in models with densities linear an infinite-dimensional parameter. then introduce novel “generalized information.” If positive, it implies (possibly irregular) when other conditions hold. impossibility results rates estimation. Three examples illustrate applicability general results. First, consider canonical example average densities. Second, show irregular median willingness pay contingent valuation studies. Finally, study discount factor and measures risk aversion nonparametric Euler equation measurement error consumption.