Abstract: Conditional likelihood methods are applied to data from removal experiments to both model capture probabilities in terms of auxillary variables, corresponding to observable individual characteristics and environmental conditions, and to estimate the size of the population. Asymptotic properties of the resulting estimators are derived and the methods are applied to a data set which include time dependent covariates. A simulation study is also given.
Key words and phrases: Covariates, log-linear models, removal method.