Bayesian Estimation of Cox Models with Non-Nested Random Effects: An Application to the Ratification of ILO Conventions by Developing Countries[Estimation bayésienne de modèles de Cox à effets aléatoires non-emboîtés : une application à la ratification des conventions de l'OIT par les pays en voie de développement]

This empirical study investigates the ratification behaviour of ILO member states. We use a multivariate hazard model for the analysis of data on the timing of ratifications of different conventions. The model accounts for two random effects, one at the country level and the other at the convention level. We use of a semi-parametric Bayesian approach, based on the partial likelihood. Our findings confirm the results of preceding studies that ratification behaviour varies substantially across member states and conventions. Furthermore, the results yield insights on the impact of unobserved heterogeneity on the ratification process.