Title: The asymptotic validity of ``standard'' FM-OLS estimation and inference in cointegrating polynomial regressions
Authors: Oliver Stypka - Technical University Dortmund (Germany)
Peter Grabarczyk - Technical University Dortmund (Germany)
Rafael Kawka - Technical University Dortmund (Germany)
Martin Wagner - University of Klagenfurt (Austria) [presenting]
Abstract: Estimation and inference in cointegrating polynomial regressions is considered which includes regressions with deterministic variables, integrated processes and their powers as explanatory variables. The stationary errors are allowed to be serially correlated and the regressors are allowed to be endogenous. The main result shows that estimating such relationships using the fully modified OLS approach developed for linear cointegrating relationships by incorrectly considering all integrated regressors and their powers as integrated regressors leads to the same limiting distribution as the fully modified type estimator developed for cointegrating polynomial regressions. A key ingredient for the main result are novel limit results for kernel weighted sums of properly scaled nonstationary processes involving scaled powers of integrated processes. Even though the simulation results indicate performance advantages of the latter estimator that are partly present even in large samples, the results drastically enlarge the usability of the former estimator as implemented in many software packages.