In order to capture observed asymmetric dependence in international financial returns, we construct a multivariate regime-switching model of copulas. We model dependence with one Gaussian and one canonical vine copula regime. Canonical vines are constructed from bivariate conditional copulas and provide a very flexible way of characterizing dependence in multivariate settings. We apply the model to returns from the G5 and Latin American regions, and document two main findings. First, we discover that models with canonical vines generally dominate alternative dependence structures. Second, the choice of copula is important for risk management, because it modifies the Value at Risk (VaR) of international portfolio returns. JEL Classification codes: C32, C35, G10.

This paper introduces new models for time series count data. The Autoregressive Conditional Poisson model ACP makes it possible to deal with issues of discreteness, overdispersion variance greater than the mean and serial correlation. A fully parametric approach i s taken and a marginal distribution for the counts is speci ed, where conditional on past observations the mean is autoregressive. This enables to attain improved inference on coe cients of exogenous regressors relative to static Poisson regression, which is the main concern of the existing literature, while modeling the serial correlation in a exible way. A v ariety of models, based on the double Poisson distribution of Efron 1986 is introduced, which in a rst step introduce an additional dispersion parameter and in a second step make this dispersion parameter time-varying. All models are estimated using maximum likelihood which makes the usual tests available. In this framework autocorrelation can be tested with a straightforward likelihood ratio test, whose simplicity is in sharp contrast with test procedures in the latent v ariable time series count model of Zeger 1988. The models are applied to the time series of monthly polio cases in the U.S between 1970 and 1983 as well as to the daily number of price change durations of :75$ on the IBM stock. A :75$ price-change duration is de ned as the time it takes the stock price to move b y at least :75$. The variable of interest is the daily number of such durations, which is a measure of intradaily volatility, since the more volatile the stock price is within a day, the larger the counts will be. The ACP models provide good density forecasts of this measure of volatility.

No abstract