A0401
Title: Smooth transition cointegrating regressions: Modified nonlinear least squares estimation and inference
Authors: Karsten Reichold - TU Wien (Austria) [presenting]
Martin Wagner - University of Klagenfurt, Bank of Slovenia and Institute for Advanced Studies, Vienna (Austria)
Abstract: Fully modified and dynamic nonlinear least squares estimators are developed for smooth transition cointegrating regressions that include deterministic and integrated variables as regressors and an integrated variable or time as a transition variable. The stationary errors are allowed to be serially correlated, and the regressors, as well as the stochastic transition variable, are allowed to be endogenous. Both estimators are shown to have the same zero-mean Gaussian mixture limiting distribution that allows for asymptotic standard inference. The theoretical analysis is complemented by a simulation study that shows that the performance advantages of the modified estimators over nonlinear least squares are comparable to the performance advantages observed in linear cointegrating regressions. Finally, the developed methodology is used to investigate potential nonlinearities of long-run US money demand.