Premium
On SETAR non‐linearity and forecasting
Author(s) -
Clements Michael P.,
Franses Philip Hans,
Smith Jeremy,
van Dijk Dick
Publication year - 2003
Publication title -
journal of forecasting
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.543
H-Index - 59
eISSN - 1099-131X
pISSN - 0277-6693
DOI - 10.1002/for.863
Subject(s) - setar , autoregressive model , econometrics , monte carlo method , interval (graph theory) , linearity , computer science , set (abstract data type) , data set , star model , statistics , mathematics , autoregressive integrated moving average , time series , engineering , combinatorics , electrical engineering , programming language
We compare linear autoregressive (AR) models and self‐exciting threshold autoregressive (SETAR) models in terms of their point forecast performance, and their ability to characterize the uncertainty surrounding those forecasts, i.e. interval or density forecasts. A two‐regime SETAR process is used as the data‐generating process in an extensive set of Monte Carlo simulations, and we consider the discriminatory power of recently developed methods of forecast evaluation for different degrees of non‐linearity. We find that the interval and density evaluation methods are unlikely to show the linear model to be deficient on samples of the size typical for macroeconomic data. Copyright © 2003 John Wiley & Sons, Ltd.