Conditional Copula Models Research Articles

Statistical testing of covariate effects in conditional copula models

In conditional copula models, the copula parameter is deterministically linked to a covariate via the calibration function. The latter is of central interest for inference and is usually estimated nonparametrically. However, in many applications it is scientifically important to test whether the calibration function is constant or not. Moreover, a correct model of a constant relationship results in significant gains of statistical efficiency. We develop methodology for testing a parametric formulation of the calibration function against a general alternative and propose a generalized likelihood ratio-type test that enables conditional copula model diagnostics. We derive the asymptotic null distribution of the proposed test and study its finite sample performance using simulations. The method is applied to two data examples.

In mixed company: Bayesian inference for bivariate conditional copula models with discrete and continuous outcomes

Conditional copula models are flexible tools for modelling complex dependence structures in regression settings. We construct Bayesian inference for the conditional copula model adapted to regression settings in which the bivariate outcome is continuous or mixed. The dependence between the copula parameter and the covariate is modelled using cubic splines. The proposed joint Bayesian inference is carried out using adaptive Markov chain Monte Carlo sampling. The deviance information criterion (DIC) is used for selecting the copula family that best approximates the data and for choosing the calibration function. The performances of the estimation and model selection methods are investigated using simulations.

Dependence Calibration in Conditional Copulas: A Nonparametric Approach

The study of dependence between random variables is a mainstay in statistics. In many cases, the strength of dependence between two or more random variables varies according to the values of a measured covariate. We propose inference for this type of variation using a conditional copula model where the copula function belongs to a parametric copula family and the copula parameter varies with the covariate. In order to estimate the functional relationship between the copula parameter and the covariate, we propose a nonparametric approach based on local likelihood. Of importance is also the choice of the copula family that best represents a given set of data. The proposed framework naturally leads to a novel copula selection method based on cross-validated prediction errors. We derive the asymptotic bias and variance of the resulting local polynomial estimator, and outline how to construct pointwise confidence intervals. The finite-sample performance of our method is investigated using simulation studies and is illustrated using a subset of the Matched Multiple Birth data.

Does Weather Matter?

We use semi-parametric bin tests, regression analyses and copula modeling techniques to identify the relationship between temperature and stock market returns. After examining 25 international stock markets, we find that the negative correlation is statistically significant in individual countries, i.e. the higher is the temperature, the lower the stock returns. However, we fail to find joint significance of temperature effects across markets after correcting for market comovement by seemingly unrelated regression. We also find negative temperature effects on returns are robust to different measures of daily temperature. Both constant-dependence and time-varying-dependence conditional copula models are employed to analyze the general dependence between temperature and stock market returns. The copula results show that the negative relation remains after controlling for autocorrelations, GARCH effects and non-normality and the dependence between temperature and stock market returns is relatively stable over time.

Frequent Turbulence? A Dynamic Copula Approach

How common and how persistent are turbulent periods? We address these questions by developing and applying a dynamic dependence framework. In order to answer the first question we estimate an unconditional mixture model of normal copulas, based on both economic and econometric justification. In order to answer the second question, we develop and estimate a hidden markov model of copulas, which allows for dynamic clustering of correlations. These models permit one to infer the relative importance of turbulent and quiescent periods in international markets. Empirically, the three most striking findings are as follows. First, for the unconditional model, turbulent regimes are more common. Second, the conditional copula model dominates the unconditional model. Third, turbulent regimes tend to be more persistent.