Approximate Confidence Interval for the Difference Binomial Proportions with Adjusted Newcombe Hybrid Score Method

Main Article Content

นพวรรณ จงสง่ากลาง
Poliny Ung
พูลพงศ์ สุขสว่าง

Abstract

The purposes of this research were to develop confidence intervals for the difference of binomial proportion with the adjusted delta method, and to compare efficiency of confidence interval between the Newcombe Hybrid Score method and the adjusted Newcombe Hybrid Score method. Research methodology was simulation data in r by the situations with sample sizes, parameters of the difference of binomial proportions, and nominal level 224 conditions (28 x 4 x 2). Then to compare the efficiency, data were analyzed about coverage probability and average width of confidence interval for each method. As a result, confidence intervals for the difference of binomial proportions with the adjusted Newcombe Hybrid Score method were lower limit and upper limit when . For comparison the efficiency for both confidence intervals, it was found that the adjusted Newcombe Hybrid Score method had a greater efficiency than the Newcombe Hybrid Score method with the following conditions: (1) both equal sample case, sample sizes were more than 25 and the parameter of the difference of binomial proportions did not exceed 0.4, and (2) unequal sample case, the first sample with more than 15, the second one with more than 50 with the parameter of the difference of binomial proportions did not exceed 0.4. Therefore, confidence intervals for each methods are appropriate depending on sample size and difference of binomial proportion.

Downloads

Download data is not yet available.

Article Details

Section
วิทยาศาสตร์กายภาพ
Author Biographies

นพวรรณ จงสง่ากลาง

วิทยาลัยวิทยาการวิจัยและวิทยาการปัญญา มหาวิทยาลัยบูรพา ตำบาลแสนสุข อำเภอเมือง จังหวัดชลบุรี 20131

Poliny Ung

วิทยาลัยวิทยาการวิจัยและวิทยาการปัญญา มหาวิทยาลัยบูรพา ตำบาลแสนสุข อำเภอเมือง จังหวัดชลบุรี 20131

พูลพงศ์ สุขสว่าง

วิทยาลัยวิทยาการวิจัยและวิทยาการปัญญา มหาวิทยาลัยบูรพา ตำบาลแสนสุข อำเภอเมือง จังหวัดชลบุรี 20131

References

[1] Hand, D.J., 2008, Statistics: A Very Short Introduction, Oxford University Press, New York, 136 p.
[2] Sasiwimonrit, K., Hirunwong, A. and Chomtee, B., 2015, Comparison of confidence interval estimation methods for the difference between two independent population proportion with small sample, Thai Sci. Technol. J. 23(2): 183-193. (in Thai)
[3] Rosenthal, J.A., 2012, Statistics and Data Interpretation for Social Work, Springer Publishing Company, New York, 512 p.
[4] Kang, S.H., 2008, New confidence interval for the difference between two proportions in two-sample correlated binary data, J. Korean Stat. Soc. 37: 175-183.
[5] Maruo, K. and Kawai, N., 2014, Confidence intervals based on some weighting functions for the difference of two binomial proportions, Stat. Med. 33: 2288-2296.
[6] Faculty of Tropical Medicine, Mahidol University, 2014, Textbook of Clinical Research, Available Source: http://www.tm.mahidol.ac.th/th/tropical-medicine-knowledge/book-clinic/Textbook-of-Clinical-Researh.php, January 16, 2018. (in Thai)
[7] Riffenburgh, R.H., 2012, Confidence Interval: Statistics in Medicine, 3rd Ed., Academic Press, Cambridge, MA.
[8] Bonett, D.G. and Price, R.M., 2012, Adjusted wald confidence intervals for a difference of binomial proportions based on paired data, J. Edu. Behav. Stat. 37: 479-488.
[9] Newcombe, R.G., 1998, Improved confidence intervals for the difference between binomial proportions based on paired data, Stat. Med. 17: 2635-2650.
[10] Agresti, A. and Caffo, B., 2000, Simple and effective confidence intervals for proportions and differences of proportions result from adding two successes and two failures, Am. Stat. 54: 280-288.
[11] Zhou, X.H., Tsao, M. and Qin, G., 2004, New intervals for the difference between two independent binomial proportions, J. Stat. Plan. Infer. 123: 97-115.
[12] Zhou, X.H. and Qin, G., 2005, A new confidence interval for the difference between two binomial proportions of paired data, J. Stat. Plan. Infer. 128: 527-542.
[13] Zou, G. and Donner, A., 2004, A simple alternative confidence interval for the difference between two proportions, Control Clin. Trials 25: 3-12.
[14] Brown, L. and Li, X., 2005, Confidence intervals for two sample binomial distribution, J. Stat. Plan. Infer. 130: 359-375.
[15] Reed, J.F.III, 2009, Improved confidence intervals for the difference between two proportions, J. Mod. Appl. Stat. Methods 8: 207-214.
[16] Fagerland, M.W., Lydersen, S. and Laake, P., 2015, Recommended confidence intervals for two independent binomial proportions, Stat. Methods Med. Res. 24: 224-254.
[17] Prendergast, L.A. and Staudte. R.G., 2014, Better than you think: Interval estimators of the difference of binomial proportions, J. Stat. Plan. Infer. 148: 38-48.
[18] Rahardja, D. and Yang, Y., 2015, Maximum likelihood estimation of a binomial proportion using one-sample misclassified binary data, Stat. Neerl. 69: 272-280.
[19] Papanicolaou, A., 2009, Taylor Approxima tion and the Delta Method, Available Source: http://web.stanford.edu/class/cme308/OldWebsite/notes/TaylorAppDeltaMethod.pdf, May 28, 2018.
[20] Ginestet, C.E., 2013, Delta Method, Available Source: http://math.bu.edu/people/cgineste/classes/ma575/p/w10_1.pdf, January 20, 2018.
[21] Ghosh, B.K., 1979, Comparison of some approximate confidence intervals for binomial parameter, J. Am. Stat. Assoc. 74: 894-900
[22] Department of Statistics, Online Programs, The Pennsylvania State University, 2018, Lesson 30: Confidence Intervals for One Mean (An Interval's Length), Available Source: https://onlinecourses.science.psu.edu/stat414/node/198, Jan 15, 2019.