ABSTRACT
This chapter deals with population size estimation in a particularly interesting case. It assumes that there is uncertainty regarding the fact that some observed individuals actually belong to the population of interest. The chapter illustrates the Scotland Drug Injectors data set discussed in Overstall and their group. The aim of this study was to estimate the number of people who inject drugs (PWID) in Scotland. The data was collected in three different time points: 2003, 2006 and 2009. The chapter details the log-linear models for possibly left-censored counts, and provides the first generalizations by discussing some simple forms of unobserved heterogeneity. It discusses choices for prior parameters, and their rationale; additionally it uses the Deviance Information Criterion (DIC) for model choice. The chapter briefs how to sample from the posterior distribution of model parameters. It illustrates several options for model specification of the Scotland Drug Injectors data set.
