European Business Schools Librarian's Group

SSE/EFI Working Paper Series in Economics and Finance,
Stockholm School of Economics

No 131: Modelling Economic Relationships with Smooth Transition Regressions

Timo Teräsvirta ()
Additional contact information
Timo Teräsvirta: Department of Economic Statistics, Postal: Stockholm School of Economics, Box 6501, 113 83 Stockholm, Sweden

Abstract: This paper has been prepared for Handbook of Applied Economic Statistics, edited by David Giles and Aman Ullah. It considers a particular class of single-equation nonlinear multivariate models called smooth transition regression (STR) models. Inference in these models, including testing linearity against STR and testing Granger noncausality, is discussed. A modelling cycle, consisting of the specification, estimation, and evaluation of these models is presented and its different stages considered in detail. Model encompassing also receives attention. Furthermore, the chapter contains a previously unpublished empirical application of the STR model to modelling UK housing price expectations. This example illustrates the workings of the modelling cycle and possible usefulness of the STR model in dynamic macroeconomic modelling.

Keywords: Causality; econometric modelling; linearity test; misspecification test; nonlinear model; structural change

JEL-codes: C20; C50

79 pages, November 1996

Download statistics

Questions (including download problems) about the papers in this series should be directed to Helena Lundin ()
Report other problems with accessing this service to Sune Karlsson ().

This page generated on 2018-02-15 23:07:16.