On Approximate Confidence Intervals for the Damage Total based on a Poisson Model

When marked count data are observed, one might be interested in making inference concerning damage proportion for a certain product. The main goal of this article is to construct approximate confidence intervals for a weighted damage proportion (which result in confidence intervals for the damage total) for a certain product based on a ‘Poisson model’ for paired count data. The true coverage of the intervals is evaluated using computer simulation and recommendations are provided for selecting the appropriate confidence intervals, which is based on the closeness of the achieved coverage of these intervals to the nominal value.