In this article we propose a multivariate dynamic probit model. Our model can be viewed as a nonlinear VAR model for the latent variables associated with correlated binary time-series data. To estimate it, we implement an exact maximum likelihood approach, hence providing a solution to the problem generally encountered in the formulation of multivariate probit models. Our framework allows us to study the predictive relationships among the binary processes under analysis. Finally, an empirical study of three financial crises is conducted.
The authors thank Lutz Kilian, Anthony Murphy, and Thomas Fomby, the editors of Advances in Econometrics for comments and discussions. We also benefited from the reaction of audiences at 12th Annual Advances in Econometrics Conference Vector Autoregressive Models: New Developments and Applications in Dallas, the 6th Method in International Finance Network Congress in Sydney, the 65th European Meeting of the Econometric Society in Oslo as well as at internal seminars at the European University Institute, the research department of the IMF, the Bundesbank-European Central Bank-Frankfurt University, the University Catholique de Louvain-CORE, the National Bank of Tunisia, the National Bank of Serbia. The usual disclaimers apply.
Candelon, B., Dumitrescu, E.-I., Hurlin, C. and Palm, F.C. (2013), "Multivariate Dynamic Probit Models: An Application to Financial Crises Mutation", VAR Models in Macroeconomics – New Developments and Applications: Essays in Honor of Christopher A. Sims (Advances in Econometrics, Vol. 32), Emerald Group Publishing Limited, Leeds, pp. 395-427. https://doi.org/10.1108/S0731-9053(2013)0000031011
Emerald Group Publishing Limited
Copyright © 2013 Emerald Group Publishing Limited