A0487
Title: Estimation and inference of panel data models with a generalized factor structure
Authors: Juan Manuel Rodriguez-Poo - Universidad de Cantabria (Spain)
Alexandra Soberon - Universidad de Cantabria (Spain)
Stefan Sperlich - University of Geneva (Switzerland) [presenting]
Abstract: A novel panel data model is introduced with a fairly general structure for the unobserved common factors, which also encompasses both traditional additive and interactive fixed effects. Under rather weak assumptions, consistent estimators are obtained that are asymptotically normal at rate root-NT for the parameters of interest, while optimal nonparametric estimators are obtained for the unspecified part. Furthermore, the statistical properties of the resulting estimators are robust to misspecification of the relationship between common factors and factor loadings. A nonparametric specification test for the crucial modeling assumption is provided. It relies on combining the methodology of conditional moment tests and nonparametric estimation techniques. Using degenerate and nondegenerate theories of U-statistics, its convergence and asymptotic distribution are shown under the null, and it diverges under the alternative at a rate arbitrarily close to root-NT. Finite sample inference is based on bootstrap. Simulations reveal an excellent performance of the methods. They are used to study the effect of the European Union emissions trading system on CO2 emission and the economy of some countries of the European Union.
