演講者簡介 : Professor Prokhorov received his Ph.D. in Economics from Michigan State University in 2006. He is currently an Associate Professor at the University of Sydney. His research interests include theoretical and applied econometrics, with applications in business analytics, finance, risk management, labour and health economics, demography and other fields.
演講摘要 : We consider estimation of stochastic frontier models with endogenous inputs and environmental variables. First, we consider modifications of standard procedures such as 2SLS and LIML to stochastic frontiers, when only inputs can be endogenous, that is we allow for dependence between x and components of the error term ε=v-u. Second, we consider models with ε=v-u^0 exp(q^' δ) and environmental variables q. The first case is nested in the second case when the input variables are correlated with v and u but q are exogenous. We allow q to be endogenous in the sense that they are not independent of v and/or u^0. We consider estimation by IV and by MLE. MLE requires specification of reduced form equations. The case that q and u^0 are dependent is difficult because we need to assume a copula and so there are computational issues associated with maximum simulated likelihood estimation.
