Lagrange Multiplier Test for both Regular and Seasonal Unit Roots

  • Park, Young-J. (Department of Computer Science and Statistics, Seoul National University) ;
  • Cho, Sin-Sup (Department of Computer Science and Statistics, Seoul National University)
  • Published : 1995.12.01

Abstract

In this paper we consider the multiple unit root tests both for the regular and seasonal unit roots based on the Lagrange Multiplier(LM) principle. Unlike Li(1991)'s method, by plugging the restricted maximum likelihood estimates of the nuisance parameters in the model, we propose a Lagrange multiplier test which does not depend on the existence of the nuisance parameters. The asymptotic distribution of the proposed statistic is derived and empirical percentiles of the test statistic for selected seasonal periods are provided. The power and size of the test statistic for examined for finite samples through a Monte Carlo simularion.

Keywords

References

  1. Statistics and Probability Letters v.16 Some tests for Unit Roots in Seasonal Time Series with Deterministic Trends Ahn,S.K.;Cho,S.
  2. Journal of Korean Statistical Society v.22 no.1 A Note on Tests for Seasonal Unit Roots in the Presence on Deterministic Trends Ahn,S.K.;Cho,S.
  3. The Annals of Statistics v.16 Limiting Distributions of Least Squares Estimates of Unstable Autoregressive Processes Chan,N.H.;Wei,C.Z.
  4. The Annals of Statistics v.7 no.5 Estimation for Autoregressive Processes with Unit Roots Hasza,D.P.;Fuller,W.A.
  5. The Annals of Statistics v.10 no.4 Testing for Nonstationary Parameter Specifications in Seasonal Time Series Models Hasza,D.P.;Fuller,W.A.
  6. Biometrika v.78 Some Lagrange Multiplier Tests for Seasonal Differencing Li,W.K.
  7. Oxford Bulletin of Economics and Statistics v.54 LM Tests for a Unit Rooy in the Presence of Deterministic Trends Schmidy,P.;Phillips,P.C.B.