Bayesian modelling of the time delay between diagnosis and settlement for Critical Illness Insurance using a Burr generalised-linear-type model

► We model the delay between diagnosis and settlement in Critical Illness Insurance. ► Data were supplied by CMI and relate to claims settled in the years 1999-2005. ► A Bayesian 3-parameter Burr generalised-linear-type model is fitted using MCMC. ► Variable selection is used to choose claim-related factors that affect the delay. ► We estimate missing diagnosis dates and allow claims to be classified accordingly.