ABSTRACT
The ratio regression approach can be extended to incorporate the information coming from the validation sample into the ratio regression model. A zero-inflated model was also considered as it appears we have a large number of zeros in addition to those predicted by the non-inflated models. The identical model might be used for the positive and validation samples, or a partly congruent model such as the parallel lines model, or two separate models such as the separate lines model. In the latter case, there is no gain in efficiency. The ratio regression approach was discussed and it could be seen how the ratio regression approach for the positive sample could be extended to include information from the validation sample, the untruncated sample including zero counts which are not observed in conventional capture-recapture settings. A zero-inflated model was considered allowing the first ratio to be particularly lower than the other ratios.
