A0291
Title: Estimation and testing for varying coefficient multidimensional panel data models: A differencing approach
Authors: Daniel Henderson - University of Alabama (United States)
Alexandra Soberon - Universidad de Cantabria (Spain) [presenting]
Christopher Parmeter - University of Miami (United States)
Abstract: An estimation method and an array of hypothesis tests are presented for varying coefficient multidimensional panel data regression models. The asymptotic distribution of the proposed nonparametric estimator is derived, and the necessary central limit theory is developed to conduct inference and to construct valid tests. The presence of multiple effects over differing dimensions requires nontrivial changes to the well-known central limit theory for U statistics. The types of inference conducted offer a diverse array of hypotheses for applied work, and test statistics are explicitly presented for some of the most important hypothesis tests. To illustrate the usefulness of the proposed suite of tests, an empirical application is provided focusing on the gravity model of international trade. It is found that the standard linear in parameters model is misspecified, suggesting the presence of modeled nonlinearities that could potentially prove useful for policy recommendations. Lastly, an appendix contains a detailed set of simulations that supports the asymptotic developments and reveals that the testing infrastructure possesses correct asymptotic size and high power.
