Proportional hazards models for current status data: application to the study of differentials in age at weaning in Pakistan

Survival analysis can be used to study the association between a set of explanatory variables and the age at which an event of interest, say weaning, occurs. As retrospectively reported age data often contain serious reporting errors, many analysts prefer to use current status data. This paper discusses two approaches to fitting survival models to current status data from the Pakistan Fertility Survey. A generalized linear model approach, using constrained maximum likelihood estimates, proves computationally expensive, so a transformation of the baseline hazard is modelled, using spline function. (Original abstract--Amended)