Abstract
Registrations in epidemiological studies suffer from incomplete ness, thus a general consensus is to use capture-recapture models. Inclusion of covariates which relate to the capture probabilities has been shown to improve the estimate of population size. The covariates used have to be measured by all the registrations. In this article, we show how multiple im putation can be used in the capture-recapture problem when some lists do not measure some of the covariates or alternatively if some covariates are unobserved for some individuals. The approach is then applied to data on neural tube defects from the Netherlands
Highlights
The estimation of the population size based on multiple incomplete lists has a long history (Chao et al, 2001, Schwarz and Seber, 1999)
Our results show that mean imputation performs well with respect to the estimate of the population size but seemingly underestimates the standard error, resulting in narrow confidence intervals
The estimate of the population size from mean imputation is similar to the estimate derived from multiple imputation because the proportion of observations with missing data is very low
Summary
The estimation of the population size based on multiple incomplete lists has a long history (Chao et al, 2001, Schwarz and Seber, 1999). In the CRC problem, Zwane and Van der Heijen (2007) considered a log-linear model to describe the multinomial probabilities among discrete (and discretized) covariates and suggested the use of the EM algorithm (Little and Rubin, 1987) for likelihood maximization when some covariates do not appear in some registrations. This approach can be used for our data but we would ignore part of the information available in the continuous covariates.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
