On unit root testing with smooth transitions

Improved critical values are calculated for Dickey–Fuller-type t ratio unit root tests against trend stationarity about non-linear trend, which is based on one deterministic smooth transition (STR) function. Simulation employs fine grid-search over both STR parameters to find accurate staring values...

Full description

Saved in:
Bibliographic Details
Published in:Computational statistics & data analysis 2006-11, Vol.51 (2), p.797-800
Main Author: Vougas, Dimitrios V.
Format: Article
Language:English
Subjects:
Deterministic smooth transition
Exact sciences and technology
General topics
Markov processes
Mathematics
Multivariate analysis
Nonlinear trend
Numerical analysis
Numerical analysis. Scientific computation
Numerical methods in probability and statistics
Probability and statistics
Probability theory and stochastic processes
Sciences and techniques of general use
Statistics
Structural change
Unit root
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Improved critical values are calculated for Dickey–Fuller-type t ratio unit root tests against trend stationarity about non-linear trend, which is based on one deterministic smooth transition (STR) function. Simulation employs fine grid-search over both STR parameters to find accurate staring values, as well as constrained optimization. In addition, two new parsimonious models are introduced. Finally, an application of the test to the log of Real per capita GNP of USA is provided.
ISSN:0167-9473
1872-7352
DOI:10.1016/j.csda.2006.07.015