Capture~Recapture Models When Both Sources

Have Clustered Observations, Journal of the American Statistical Association, June 1986, Vol. 81, # 394, pp. 347~353, and Proceedings of the American Statistical Association, Section on Survey Research Methods, 1984.

Project details

The model most commonly used in capture-recapture estimation of populations sizes assumes that individuals are missed or captured in each source independently of all other individuals in the population. However, this assumption may be inadequate for some studies: certain research designs would lead naturally to clustering of misses in the enumeration of a population. This paper will develop a model that will describe the clustering of misses; it will describe biases inherent in the traditional dual system estimator when the model involving clustering holds, and finally, the paper will outline the use of the EM algorithm to estimate the total population size.

-Cowan, Charles D. and Malec, Donald J.

Back to Publications

Related Projects