Seminars 2008 — Abstracts
Friday, December 5
Speaker:
Tilak Abeysinghe,
Singapore
Title:
Gaussian tests for unit roots and cointegration with more power
Abstract: Non-standard distributions, size distortions and extremely low power are well-known
problems of the unit root tests that are currently in use. In this paper we
use a mixed-frequency regression technique to develop a test primarily for cointegration
under the null of stationarity. This MA unit root test, based on a variance-difference,
takes the testing back to the normal distribution and offers a way to increase
power without having to increase the sample size substantially. Monte Carlo
simulations show that the test offers substantial gains in power against near-null
alternatives in moderate size samples. Although the null of stationarity is
the research line to be pursued, for comparison we also consider an extension
of the procedure to test for an AR unit root. An empirical exercise illustrates
the usefulness of the test.
