|
Organizers |
Bayesian Unit Root Test for a Time Series Model with Trend Approximated by Linear Spline Function
by
Jitendra Kumar
Department of Mathematics & Statistics, Allahabad Agricultural Institute-Deemed University, Allahabad, UP-211007, INDIA
Coauthors: Anoop Chaturvedi
Department of Statistics, University of Allahabad, Allahabad India-211002
The present paper considers Bayesian unit root test for a time series model with non-linear trend, approximated by linear spline. The beauty of approximating the non-linear trend with linear spline function is that the time trend vanishes under the unit root hypothesis, unlike the approximation by polynomial in which the trend remains present under the unit root hypothesis. The posterior odds ration has been derived under appropriate prior assumption.
Date received: January 24, 2007
Copyright © 2007 by the author(s). The author(s) of this document and the organizers of the conference have granted their consent to include this abstract in Atlas Conferences Inc. Document # cata-46.