Freedman will work on questions about the foundations of statistics: how to characterize prior beliefs, and when do such beliefs lead to good statistical procedures; how to look at high dimensional data; what happens to standard statistical procedures under nonstandard conditions; how to test statistical models. These questions include the following topics: partial exchangeability and sufficiency; the consistency of Bayes estimates; asymptotic normality of posterior distributions; the bootstrap technique for estimating standard errors; the application of regression models to social science data.
