Panel data specifications in nonparametric kernel regression: an application to production functions

Tomasz Gerard Czekaj, Arne Henningsen

Abstract

We discuss nonparametric regression models for panel data. A fully nonparametric panel data specification that uses the time variable and the individual identifier as additional (categorical) explanatory variables is considered to be the most suitable. We use this estimator and conventional parametric panel data estimators to analyse the production technology of Polish crop farms. The results of our nonparametric kernel regressions generally differ from the estimates of the parametric models but they only slightly depend on the choice of the kernel functions. Based on economic reasoning, we found the estimates of the fully nonparametric panel data model to be more reliable.
Original languageEnglish
Place of PublicationFrederiksberg
PublisherDepartment of Food and Resource Economics, University of Copenhagen
Pages1-54
Number of pages54
Publication statusPublished - 2013
SeriesIFRO Working Paper
Number2013/5

Fingerprint

Dive into the research topics of 'Panel data specifications in nonparametric kernel regression: an application to production functions'. Together they form a unique fingerprint.

Cite this