This event is from the archives of The Notice Board. The event has already taken place and the information contained in this post may no longer be relevant or accurate.
"ENDOGENEITY IN STOCHASTIC FRONTIER MODELS: LIKELIHOOD INSTRUMENTAL VARIABLE AND COPULA APPROACH WHEN NO ADDITIONAL DATA ARE AVAILABLE"
Dr. KIEN C. TRAN will deliver this seminar on Friday, April 10 from 2:30 - 4:00pm in C-630
Abstract: This paper considers two alternative estimation procedures for estimating stochastic frontier models with endogenous regressors when no external instruments are available. In the first procedure, we use Lewbel's (1997) method to construct the instruments using higher order moment of the available data which are then used in the likelihood type GMM estimation based on the moment conditions of the likelihood scores. The second procedure which does not require the use of instruments, is based on the copula approach to directly model the correlation between the endogeous regressors and the composed errors. Monte Carlo simulations are used to assess and compare the finite sample performances of the proposed estimation procedures. An application with real data of U.S. banks is further presented to demonstrate the usefulness of the proposed methodologies.
Contact:
Merle Christie | christie@uleth.ca | (403) 329-2518