Transform MCMC Schemes for Sampling Intractable Factor Copula Models

In financial risk management, modelling dependency within a random vector X is crucial, a standard approach is the use of a copula model. Say the copula model can be sampled through realizations of Y having copula function C: had the marginals of Y been known, sampling X⁽ⁱ⁾, the i-th component of X, would directly follow by composing Y⁽ⁱ⁾ with its cumulative distribution function (c.d.f.) and the inverse c.d.f. of X⁽ⁱ⁾. In this work, the marginals of Y are not explicit, as in a factor copula model. We design an algorithm which samples X through an empirical approximation of the c.d.f. of the Y-marginals. To be able to handle complex distributions for Y or rare-event computations, we allow Markov Chain Monte Carlo (MCMC) samplers. We establish convergence results whose rates depend on the tails of X, Y and the Lyapunov function of the MCMC sampler. We present numerical experiments confirming the convergence rates and also revisit a real data analysis from financial risk management.