CMStatistics 2018: Start Registration
View Submission - CFE
A0420
Title: A nested copula duration model for competing risks with multiple spells Authors:  Ralf Wilke - Copenhagen Business School (Denmark) [presenting]
Enno Mammen - University of Heidelberg (Germany)
Ming Sum Simon Lo - City University of Hong Kong (Hong Kong)
Abstract: A copula graphic estimator for the competing risks duration model with multiple spells is presented. It is a general nested copula model that allows for different degrees of dependence between risks and spells and therefore breaks up an implicit restriction of popular duration models such as multivariate mixed proportional hazards. It is shown that the dependence structure between spells is identified and can be estimated, in contrast to the dependence structure between competing risks. Thus, by allowing these two components to differ, the model is not identified. This is an important finding related to the general identifiability of competing risks models. Various features of the model are investigated by simulations and its practicality is illustrated by an application to unemployment duration data.