Maximum likelihood estimation for the proportion difference of two-sample binomial data subject to one type of misclassification. Issue 8 (17th November 2019)

• Record Type: Journal Article
• Authors: Rahardja, Dewi; Wu, Han; Zhang, Zhiwei; Tiedt, Andrew D.
• Is Part Of: Journal of statistics & management systems. Volume 22:Issue 8(2019)
• Page Start: 1365
• Page End: 1379
• Publication Date: 2019-11-17
• Subjects: 62-07 -- 62F10 -- 62H12 -- 65C05 -- 68W01; Misclassification -- identifiability -- binary data; Statistics -- Periodicals; Mathematical models -- Periodicals; Mathematical models; Statistics; Periodicals; 519.5
• DOI: 10.1080/09720510.2019.1606319
• Languages: English
• ISSNs: 0972-0510

Abstract

Abstract: In this manuscript, we derived three likelihood-based interval estimation methods using a closed-form algorithm for the difference of two independent binomial proportion parameters with one type of misclassification. We acquired an identifiable model by using a double-sampling scheme. We also employed simulations to examine the robustness of our three likelihood-based interval estimation methods and summarize that our modified Wald method implemented to new data with Agresti-Coull type of adjustment performs well and has nominal coverage probabilities. This method was adapted to traffic data for an illustration.