Estimation and Inference in Unstable Nonlinear Least Squares Models (Final)

In this thesis, we extend Bai and Perron's (1998, Econometrica, pp. 47-78) method fordetecting multiple breaks to nonlinear models. To that end, we consider an unstable univariatenonlinear least squares (NLS) model with a limited number of parameter shifts occurring atunknown dates. In our framework, the break-dates are simultaneously estimated with theparameters via minimization of the residual sum of squares. Using nonlinear asymptotic theory,we derive the asymptotic distributions of both break-point and parameter estimates and proposeseveral instability tests. We also present simulation results that validate our procedure. Ourmethod is useful for estimating and testing nonlinear macroeconomic models with multipleunknown breaks. As an empirical illustration, we explore the relationship between our modeland smooth transition models in the context of a US interest rate reaction function. Unlikeprevious studies, our model can nest nonlinearities and breaks. We provide evidence for at leasttwo breaks while allowing for smooth transition within each regime, before and after a break.