A Multiple Unit Roots Test Based on Least Squares Estimator

  • Shin, Key-Il (Department of Statistics, Hankuk University of Foreign Studies)
  • Published : 1999.03.01

Abstract

Knowing the number of unit roots is important in the analysis of k-dimensional multivariate autoregressive process. In this paper we suggest simple multiple unit roots test statistics based on least squares estimator for the multivariate AR(1) process in which some eigenvalues are one and the rest are less than one in magnitude. The empirical distributions are tabulated for suggested test statistics. We have small Monte-Calro studies to compare the powers of the test statistics suggested by Johansen(1988) and in this paper.

Keywords

References

  1. Journal of the American Statistical Association v.74 Distribution of the Estimators for Autoregressive Time Series with a Unit Root Dickey, D. A.;Fuller, W. A
  2. Econometrica v.55 no.2 Co-Integration and Error Correction : Representation, Estimation, and Testing Engle, R. F.;Granger, C. W. I
  3. The Annals of Statistics v.17 Testing for a Unit Root Nonstationarity in Multivariate Autoregressive Time Series Fountis, N. G.;Dickey, D. A
  4. Introduction to Statistical Time Series Fuller, W. A
  5. Time Series Analysis Hamilton, J. D.
  6. Journal of Economic Dynamics and Control v.12 Statistical Analysis of Cointegration Vectors Jahansen, S
  7. The American Statistician v.48 no.1 A Drunk and Her Dog : An Illustrations of Cointegration and Error Corrction Murray, M. P
  8. Review of Economic Studies v.53 Multiple Time Series Regression with Integrated Processes Phillips, P. C. B.;Durlauf, S. N
  9. Matrix Algebra Useful for Statistics Seal, S. R