Bivariate Test for Testing the EQUALITY of the Average Areas under Correlated Receiver Operating Characteristic Curves (Test for Comparing of AUC’s of Correlated ROC Curves)

D. M. Senaratna; M.R. Sooriyarachchim; N. Meyen

American Journal of Applied Mathematics and Statistics. 2015, 3(5), 190-198
DOI: 10.12691/AJAMS-3-5-3

Original Research

Bivariate Test for Testing the EQUALITY of the Average Areas under Correlated Receiver Operating Characteristic Curves (Test for Comparing of AUC’s of Correlated ROC Curves)

D. M. Senaratna^1,, M.R. Sooriyarachchim¹ and N. Meyen¹

¹Department of Statistics, University of Colombo, Colombo 3, Sri Lanka

Pub. Date: October 16, 2015

Full Text PDF

Cite this paper

D. M. Senaratna, M.R. Sooriyarachchim and N. Meyen. Bivariate Test for Testing the EQUALITY of the Average Areas under Correlated Receiver Operating Characteristic Curves (Test for Comparing of AUC’s of Correlated ROC Curves). American Journal of Applied Mathematics and Statistics. 2015; 3(5):190-198. doi: 10.12691/AJAMS-3-5-3

Abstract

Methodology developed for comparing correlated ROC curves are mainly based on nonparametric methods. These nonparametric methods have several disadvantages. In this paper the authors propose an asymptotic bivariate test for comparing pairs of AUCs for independent data based on the Dorfman and Alf maximum likelihood approach. The properties of the test are examined by using simulation studies. The method is illustrated on an example of angiogram results from Sri Lanka. The test applied to the example found that there was a significant difference in the predictive power of three different cut-offs examined.

Keywords

bivariate test, Receiving Operating Characteristic (ROC) curve, Area under the Curve (AUC), Beta Distribution, Angiogram, Cardiac Stress Test (CST)

Copyright

This work is licensed under a Creative Commons Attribution 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

References

[1]	Berwick V, Cheek L, Ball J. Statistics review 9: One-way analysis of variance. Critical Care 2004; 8(2) : 130-136.

[2]	Bolland K, Sooriyarachchi MR, Whitehead J. Sample size review in a head injury trial with ordered categorical responses. Statistics in Medicine 1998; 17(24): 2835-2847.

[3]	Bonferroni CE. Il Calcolo Delle Assi curazioni Su Gruppi Di Test. In studi Onore Del Professore Salvatore Ortu Carboni. Rome, Italy. 1935a; 13-60.

[4]	Carpenter JR, Kenward MG. Missing data in clinical trials a practical guide. Birmingham: National Health Service Co-ordinating Centre for Research Methodology. 2008. from www.missingdata.org.uk, accessed 15 December 2009.

[5]	Casella, G & Berger, R. L. “Statistical Inference.” Duxbury Press. Second Edition, 2002.

[6]	Davey A, Savla J. Statistical Power Analysis with missing data. Taylor and Francis, New York, 2010.

[7]	DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometric. 1988, 44(3), 837-45.

[8]	Dorfman DD, Alf E. Maximum likelihood estimation of parameters of signal detection theory and determination of Confidence intervals-rating method data. Journal of Mathematical Psychology 1969 ; 6(3): 487-496.

[9]	Dorfman, D. D. & Berbaum, K. S. “Degeneracy and discrete reciever operating characteristic rating data.” Academic Radiology, 2(10), pp. 907-915, 1995.

[10]	Grey, D. M. & Morgan, B. T. “Some aspects of ROC curve fitting: Normal and Logistic models.”Journal of Mathematical Psychology,Volume 9, pp. 128-139, 1972.

[11]	Hanley JA, McNeil BJ. The meaning and Use of the Area under a Receiver Operating Characteristic (ROC) curve. Radiology 1982 ; 143(1) : 29-36

[12]	Hanley JA, McNeil BJ. A method of comparing the Areas under Receiver Operating Characteristic Curves Derived from the same cases. Radiology 1983 ; 148(3): 839-843.

[13]	Hanley JA. The robustness of the “binormal” assumptions used in fitting ROC curves. Medical Decision Making. 1988; 8(3), 197-203.

[14]	Honghu-Liu, Li G, William G. Cumberland and Tongtong Wu .Testing Statistical Significance of the Area under a Receiving Operating Characteristics Curve for Repeated Measures Design with Bootstrapping Journal of Data Science 3(2005), 257-278

[15]	Hosmer DW, Lemeshow S. Applied logistic regression: Wiley Series in probability and statistics. 2000.

[16]	Hotelling H. Multivariate Quality Control. In C. Eisenhart, M.W. Hastay and W.A.Wallis, eds. Techniques of Statistical Analysis. New York: McGraw-Hill, 1947.

[17]	Krzanowski WJ, Hand DJ. ROC curves for continuous data. CRC/Chapman and Hall; 2009.

[18]	Leeuwen MV, Zweers EJK, Opmeer BC, Ballegooie EV, Brugge HGT, Valk HWD. Comparison of Accuracy Measures of Two Screening Tests for Gestational Diabetes Mellitus. Diabetes Car. 2007; 30(11), 2779-2784.

[19]	Little RJA, Rubin DB. Statistical Analysis with missing data. New York: Willey; 1987.

[20]	Mackassy SA, Provost F. Confidence Bands for ROC curves. CeDER working paper 02-04, Stern School of Business, New York University, Ny, NY 10012; 2004.

[21]	MacMillan NA, Rotello CM, Miller JO. The sampling distributions of Gaussian ROC statistics. Perception and Psychophysics. 2004 ; 66(3) : 406-421.

[22]	Mardia KV, Kent JT, Bibby JM. Multivariate Analysis, Academic Press, 1979.

[23]	Mehta K, Rustagi M, Kohil S, Tiwari S. Implementing Multiple Imputation in an automatic Variable Selection Scenario, Proceedings of the NESUG conference, November, 2007, Blatimore, Maryland, USA; 1997.

[24]	Metz CE. Basic principles of ROC analysis. Seminars in Nuclear Medicine. 1978; 8(4), 283-298.

[25]	Metz CE, Herman BA, Roe CA. Statistical Comparison of Two ROC-curve estimates obtained from partially-paired datasets. Medical Decision Making 1998a ; 18 (1): 110-121.

[26]	C. E. Metz, P Wang, H.B. Kronman (1984). A New Approach for Testing the Significance of Differences Between ROC Curves Measured from Correlated Data. Information Processing in Medical Imaging, 432-445.

[27]	Obuchowski NA. Fundament of Clinical Research for Radiologists. American Journal of Roentgenology. 2005 ; 184 (2) : 364-372.

[28]	Park SH, Goo JM, Jo C. Receiver Operating Characteristic (ROC) curve: Practical review for radiologists. Korean Journal of Radiology. 2004 ; 5(1) : 11-18

[29]	Pepe M S. The statistical evaluation of medical Tests for classification and prediction. New York: Oxford University Press; 2003.

[30]	Royston P, Babiker A. A menu-driven facility for complex sample size calculation in randomized controlled trials with a survival or a binary outcome. The Stata Journal 2002 ; 2(2) : 151-163.

[31]	Royston, P. Multiple imputation of missing values: Update. The Stata Journal 2005 ; 5(2): 188-201.

[32]	Schafer JL. Analysis of Incomplete Data.: Chapman and Hall, 1997.

[33]	Senaratna, D. M. and. Sooriyarachchi, M.R., Determining the Sensitivity and Specificity of a Alternative Test as a Diagnostic for its Gold Standard in the presence of Severe Missingness. Journal of the National Science Foundation, Sri Lanka, 2012 40(4) 321-331.

[34]	Spritzler J, Degruttola VG, Pei L. Two-Sample Tests of Area-Under-the-Curve in the Presence of Missing Data. Int J Biostat. 2008 Jan, 2008; 17: 4(1): Article1.

[35]	Vergara IA, Normbuena T, Ferrada E, Slater AW, Melo F. StAR: a simple tool for the statistical comparison of ROC curves. BMC Bioinformatics. 2008; 9:265 Open Access.

[36]	Wishart J. The generalized Product Moment Distribution in Samples from a Normal Multivariate Population, Biometrika 1928; 20A (1-2): 32-52.

[37]	https://www.heart.org/idc/groups/heart-public/@wcm/@hcm/documents/downloadable/ucm_317626.pdf. Cardiac Catheterization and Angiogram. American Heart Association (2010). Downloaded on 08th January 2015.