A competing risk study of menarcheal age distribution based on non-recall current status data.

In many cross-sectional studies, the chances that an individual will be able to exactly recall the event are very low. The possibility of recalling the exact time as well as the cause of occurrence of an event usually decreases as the gap between event and monitoring time increases. This gives rise to non-recall current status data. In this article, an efficient approach to deal with such non-recall current status data is established in a competing risk set up. In the classical method, a nested Expectation-Maximization technique is worked out for the estimation purpose and the information matrix is evaluated using the missing information principle. In the Bayesian paradigm, point and interval estimates are obtained using the Gibbs sampling algorithm. A recent anthropometric study data containing the menarcheal status of girls and age at menarche is analyzed using the considered methodology.
Competing Interests: No potential conflict of interest was reported by the author(s).
(© 2022 Informa UK Limited, trading as Taylor & Francis Group.)