Semiparametric estimation of nonlinear panel-data models


Update content

Title Semiparametric estimation of nonlinear panel-data models
Period 07 / 2011 - 07 / 2015
Status Current
Dissertation Yes
Research number OND1342447
Data Supplier NWO


The use of panel data in economic applications steadily increases, but there is limited knowledge regarding estimation of nonlinear fixed-effect models. Most estimators are designed for particular models and analogs of general semiparametric estimation available for cross-sectional data do not practically exist. To improve upon this, a method for correcting bias of semiparametric estimators is first proposed, based on the indirect inference principle. Next, this bias correction technique is used to adapt existing semiparametric estimators of single-index models to nonlinear fixed-effect models. Finally, a new rank-correlation estimator is proposed for panel models with limited dependent response variables.

Related organisations

Related people

Project leader Dr. P. Cizek
Doctoral/PhD student J. Lei

Go to page top
Go back to contents
Go back to site navigation