Open to: Academic | Student
Admission: UK postgraduate students: £60; UK academics: £120; other delegates: £440
Ticketing: Ticketed and Pre-booking essential

Speaker information

Jeffrey Woodbridge, MSU

PEPA Training Course

This course will focus on the use of control function methods for policy analysis. The term "control function" will be used broadly, so that traditional endogeneity problems are covered along with the Chamberlain-Mundlak approach to correlated random effects panel data models. Panel data models with both endogeneity and unobserved heterogeneity will be covered. The special problems with discrete endogenous explanatory variables, and the limitations for the control function approach for nonlinear models with such variables, will be covered. Other topics include difference-in-differences estimation - both parametric and some nonparametric approaches - and the issues that arise with clustered data. We will also study methods of dealing with missing data, including imputation and probability weighting. Time permitting, approaches for handling unbalanced panels will be covered.

Contact

Kylie Groves
020 7291 4821 |

Links

More information