Get access

TESTING THE STRUCTURE OF CONDITIONAL CORRELATIONS IN MULTIVARIATE GARCH MODELS: A GENERALIZED CROSS-SPECTRUM APPROACH

Authors

  • Nadine McCloud,

    1. University of the West Indies at Mona, Jamaica; Cornell University, U.S.A., and Xiamen University, China
    Search for more papers by this author
  • Yongmiao Hong

    1. University of the West Indies at Mona, Jamaica; Cornell University, U.S.A., and Xiamen University, China
    Search for more papers by this author
    • We are sincerely grateful to the editor (Frank Schorfheide) and three anonymous referees for their invaluable critiques that helped to craft this article. Also, we thank Wolfgang Härdle, Tae-hwy Lee, Anton Schick, and seminar participants at the 2008 Xiamen University-Humboldt University Joint Workshop on Economics and Finance, 2008 Far Eastern and South Asian Meeting of the Econometric Society, and the Statistics Seminar Series in Department of Mathematical Sciences at State University of New York at Binghamton for their comments and suggestions. Hong thanks the National Science Foundation of China for its support via the Overseas Outstanding Youth Grant. McCloud thanks the Department of Economics and Graduate Student Organization at the State University of New York at Binghamton and the Department of Economics and the Dean of the Faculty of Social Sciences at the University of the West Indies at Mona, and Wang Yanan Institute for Studies in Economics at Xiamen University for travel support. All errors are our own. Please address correspondence to: Yongmiao Hong, Department of Economics, Cornell University, Ithaca, NY 14853. Phone: 607 255 5130; Fax: 607-255-2818; E-mail: yh20@cornell.edu.


  • Manuscript received October 2008; revised January 2010.

Abstract

We introduce a class of generally applicable specification tests for constant and dynamic structures of conditional correlations in multivariate GARCH models. The tests are robust to the presence of time-varying higher-order conditional moments of unknown form and are pure significance tests. The tests can identify linear and nonlinear misspecifications in conditional correlations. Our approach does not necessitate a particular parameter estimation method and distributional assumption on the error process. The asymptotic distribution of the tests is invariant to the uncertainty in parameter estimation. We assess the finite sample performance of our tests using simulated and real data.

Ancillary