In recent years, stationary time series models based on copula functionsbecame increasingly popular in econometrics to model nonlineartemporal and cross-sectional dependencies. Within these models, weconsider the problem of testing the goodness-of-fit of the parametricform of the underlying copula. Our approach is based on a dependentmultiplier bootstrap and it can be applied to any stationary, stronglymixing time series. The method extends recent i.i.d. results by Kojadinovic,Yan and Holmes [I. Kojadinovic, Y. Yan and M. Holmes,Fast large sample goodness-of- fit tests for copulas, Statistica Sinica21 (2011), 841{871] and shares the same computational benefits comparedto methods based on a parametric bootstrap. The finite-sampleperformance of our approach is investigated by Monte Carlo experimentsfor the case of copula-based Markovian time series models.
展开▼