Analysis of an Extended Batch Marking Experiment Using Estimating Equations

Estimating equations give a flexible method of inference when the likelihood is either intractable or is not fully specified. Here we consider a batch marking experiment where the full likelihood is complex and without resorting to devices such as the computationally intensive expectation-maximization (EM) algorithm is intractable. However, a pseudo-likelihood that yields tractable estimating equations may be easily constructed and its effectiveness is demonstrated via simulation studies. The methodology was applied to study the abundance of Oriental weatherloach in a waterbody in South-Eastern Australia. The programs and data set used in this article are available in the online supplements.