Output details
11 - Computer Science and Informatics
University of East Anglia
Multivariate logit copula model with an application to dental data
<13>Biostatistics is a prime area of application for copula models, as the traditional assumption of multivariate normality is usually invalid in this context. We introduce a multivariate copula-based logit model to examine the associations among multivariate binary responses. Our model has some interesting features: it provides complete inference, i.e. maximum likelihood estimation and calculation of joint and conditional probabilities, it allows for a flexible positive dependence structure, the use of covariate functions for the dependence parameters is straightforward, and its estimates are synthesized from different candidate copula models by adopting a novel model averaging approach on copula modelling.