To tackle the Identifiability issue and to Estimate the Misclassification Parameters
Improving Estimation of Misclassification Parameters in Cluster Randomization Studies
Keywords:
cluster randomization studies, misclassification parameters, complementary poison model, interval estimators, binomial misclassification rates, wald statistic, score statistic, profile log-likelihood statistic, semiparametric testing, model identificationAbstract
Cluster randomization studies have become more common in place of traditional trials thatrandomly assign participants one at a time when this method is impractical for theoretical, ethical, orpractical reasons. In the setting of a complementary poison model with potentially misclassified data, weevaluate three interval estimators for binomial misclassification rates: one based on the wald statistic,another on the score statistic, and a third on the profile log-likelihood statistic. As a result of itsimproved power and lower type I error, the redesigned test comes highly recommended. Semiparametrictesting of misclassification estimates Information on the parameters employed in g (x*, z) that underlieparametric models, misclassification, and model and identification-related problemsPublished
2022-03-01
How to Cite
[1]
“To tackle the Identifiability issue and to Estimate the Misclassification Parameters: Improving Estimation of Misclassification Parameters in Cluster Randomization Studies”, JASRAE, vol. 19, no. 2, pp. 129–136, Mar. 2022, Accessed: Sep. 19, 2024. [Online]. Available: https://ignited.in/index.php/jasrae/article/view/13804
Issue
Section
Articles
How to Cite
[1]
“To tackle the Identifiability issue and to Estimate the Misclassification Parameters: Improving Estimation of Misclassification Parameters in Cluster Randomization Studies”, JASRAE, vol. 19, no. 2, pp. 129–136, Mar. 2022, Accessed: Sep. 19, 2024. [Online]. Available: https://ignited.in/index.php/jasrae/article/view/13804