This report grew out of a study of methods of estimating the methods of estimating the monthly probability of conceiving (fecundability) of a group of women (Sheps (1964; 1965), but applies also to other risks that operate over a period of time, such as risks of dying or of acquiring a disease. Because the exposure-time estimator that expresses the findings as the number of events occurring per person-year of exposure is widely used in important applications, this paper focuses particularly on the characteristics of that index in defined circumstances.
Abstract