Improving Likelihood-Ratio-Based Confidence Intervals for Threshold Parameters in Finite Samples

22 Pages Posted: 27 Mar 2014 Last revised: 13 Aug 2017

See all articles by Luiggi Donayre

Luiggi Donayre

University of Minnesota - Duluth

Yunjong Eo

The University of Sydney - School of Economics

James Morley

University of Sydney

Date Written: August 10, 2017

Abstract

Within the context of threshold regressions, we show that asymptotically-valid likelihood-ratio-based confidence intervals for threshold parameters perform poorly in finite samples when the threshold effect is large. A large threshold effect leads to a poor approximation of the profile likelihood in finite samples such that the conventional approach to constructing confidence intervals excludes the true threshold parameter value too often, resulting in low coverage rates.

We propose a conservative modification to the standard likelihood-ratio-based confidence interval that has coverage rates at least as high as the nominal level, while still being informative in the sense of including relatively few observations of the threshold variable. An application to thresholds for U.S. industrial production growth at a disaggregated level shows the empirical relevance of applying the proposed approach.

Keywords: Threshold regression; Inverted likelihood ratio; Confidence Interval; Finite-sample inference

JEL Classification: C13, C20

Suggested Citation

Donayre, Luiggi and Eo, Yunjong and Morley, James, Improving Likelihood-Ratio-Based Confidence Intervals for Threshold Parameters in Finite Samples (August 10, 2017). Available at SSRN: https://ssrn.com/abstract=2415224 or http://dx.doi.org/10.2139/ssrn.2415224

Luiggi Donayre

University of Minnesota - Duluth ( email )

1318 Kirby Drive
Duluth, MN 55812
United States
2187268463 (Phone)

Yunjong Eo (Contact Author)

The University of Sydney - School of Economics ( email )

Social Sciences Building (A02)
Sydney, NSW 2006
Australia

James Morley

University of Sydney ( email )

Rm 370 Merewether (H04)
Sydney, NSW 2006 2008
Australia

HOME PAGE: http://https://sites.google.com/site/jamescmorley/

Register to save articles to
your library

Register

Paper statistics

Downloads
82
Abstract Views
585
rank
303,477
PlumX Metrics