臺灣博碩士論文加值系統

　　近幾年來，對於ROC曲線指標，有愈來愈多的文獻開始對於相似性及非劣性檢定方法提出相關的研究。例如：在醫學上一種新的非侵入性診斷方法與一種已被認定為標準的侵入性診斷方法做比較，也許在診斷的效率上並不一定相等，當經過相似性及非劣性的檢定，得到新方法與標準的診斷方法是相似的，若新方法在成本、使用方便性及安全性的考量下較優於標準方法時，將以新方法取代標準的方法。而本篇論文主要是利用無母數法、標準化差評估法、Kolmogorov-Smirnov法及拔靴法等四種方法評估相似性及非劣性的檢定方法，並以模擬結果比較這四種方法在常態分配假設、指數分配假設下的優劣。

　In recent years, for ROC curve index, more and more issues have focused on the equivalence or non-inferiority test. For example, in comparing diagnostic efficacy of a non-invasive alternative diagnostic procedure to an invasive method. If the non-invasive alternative procedure is equivalent to the invasive method, we may use the non-invasive alternative diagnostic procedure because of its easy administration, its better safety profile or its reduced cost. In this paper, the equivalence/non-inferiority tests based on four methods, the non-parametric method, method based on the standardized difference, method based on the Kolmogorov-Smirnov statistic, and bootstrap method are compared. A simulation study was conducted to empirically investigate the size and power of four methods for various combinations of distributions.

Contents
Chapter 1 Introduction 1
1.1 The ROC Curve 2
1.2 The Area under the ROC Curve (AUROC) 3
1.3 Tests for Equivalence or Non-inferiority 5

Chapter 2 Current Statistical Methods for Equivalence/Non-inferiority Test 7
2.1 Equivalence Method Based on Two AUROCs 7
2.1.1 Estimation of Areas under Correlated ROC curves 7
2.1.2 Equivalence and Non-inferiority Test Based on Two AUROCs 10
2.2 Equivalence Method Based on the Standardized Difference 11
2.1.1 Estimation of Areas under Correlated ROC curves 11
2.1.2 Equivalence and Non-inferiority Test Based on the Standardized Difference 13

Chapter 3 Proposed Methods 16
3.1 Method Based on Kolmogorov-Smirnov Statistics 16
3.1.1 Analysis of Bioequivalence Under Two Correlated Kolmogorov-Smirnov Statistics 16
3.1.2 Equivalence and Non-inferiority Test Based on the K-S Statistics 18
3.2 The Bootstrap Method 19
3.2.1 Bootstrap Procedure 20
3.2.2 Testing Hypothesis 21
Chapter 4 Simulation Study 22
4.1 Simulation Process and Parameter Combinations 22
4.2 Simulation Results 26


Chapter 6 Conclusion 30

Reference 31

Appendix A 34

Appendix B 36

Appendix C 41

References
[1] Blackwelder, W. C., Proving the null hypothesis in clinical trials, Controlled Clinical Trials, Vol.3, 345-353, 1982.
[2] Chan, I. S. F., Proving non-inferiority or equivalence of two treatments with dichotomous
endpoints exact methods, Statistical Methods in Medical Research, Vol.12, 37-56,
2003.
[3] Chen, B. C., On the equivalence and non-inferiority for the areas of two ROC curves,
Unpublished Master Thesis of the Department of the Statistics of National Cheng Kung
University, 2003. (http://etdncku.lib.ncku.edu.tw/ETD-db/)
[4] DeLong, E., DeLong, D., and Clarke-Pearson, D., comparing the areas under two or
more correlated receiver operation characteristic curves: A nonparametric approach,
Biometrics, Vol.44, 837-845, 1988.
[5] Efron, B. and Tibshirani, R. J., An Introduction to the Bootstrap, Chapman and Hall,
U.S.A., 1993.
[6] Farrington, C. P., and Manning, G., Test statistics and sample size formulae for
comparative binormal trials with null hypothesis of non-zero risk difference of
non-inferiority relative risk, Statistics in Medicine, Vol.9, 1447-1454, 1990.
[7] Gendenko, B. V., Tests of Homogeneity of Probability Distributions in Two Independent Samples (in Russian), Dokl. An SSSR, 80, 525-528, 1954
[8] Hodges, The Significance Probability of the Smirnov Two-Sample Test, Arkiv for Matematik, 3, 469-486, 1959
[9] Hoeffding, W., A class of statistics with asymptotically normal distribution, Annals of
Mathematical Statistics, Vol.19, 293-325, 1948.
[10] Hsueh, H. M., Liu, J. P. and Chen, J. J., Unconditional exacts tests for equivalence or
non-inferiority for paired binary endpoints, Biometrics, Vol.57, 478-483, 2001.
[11] Liao, J., An improved concordance correlation coefficient, Pharmaceutical Statistics,
Vol.2, 253-261, 2003.
[12] Liu, J. P., Hsueh, H. M., Hsieh, E., and Chen, J. J., Tests for equivalence or
non-inferiority for paired binary data, Statistics in Medicine, Vol.21, 231-245, 2002.
[13] Lu, Y., and Bean, J. A., On the sample size for one-sided equivalence of sensitivities
based upon McNemar’s test, Statistics in Medicine, Vol.14, 1831-1839, 1995.
[14] Luchenbruch, P. A., and Lynch, C. J., Assessing screening tests: extensions of
McNemar’s test, Statistics in Medicine, Vol.17, 2207-2217, 1998.
[15] Masaryk, A. M., Ross, J. S., DiCello, M. C., Modic, M. T., Paranandi, L., and Masaryk,
T. J., 3DFT MR angiography of the carotid bifurcation: Potential and limitations as a
screening examination, Radiology, Vol.179, 797-804, 1991.
[16] Moran, P. A. P., Testing for correlation between non-negative variates, Biometrika,
Vol.54, 385-394, 1967.
[17] Nam, J-m., Establishing equivalence of two treatments and sample size requirements
in matched-pairs design, Biometrics, Vol.17, 1422-1430, 1997.
[18] Obuchowski, N. A., Nonparametric analysis of clustered ROC curve data, Biometrics,
Vol. 53, 567-578, 1997.
[19] Smirnov, N. V., On the estimation of the discrepany between empirical curves of distribution for two independent samples. Bull. Moscow Univ., 2, 3-16, 1939
[20] Obuchowski, N. A., Nonparametric analysis of clustered ROC curve data, Biometrics,
Vol. 53, 567-578, 1997.
[21] Obuchowski, N. A., and McClish, D. K., Sample size determination for diagnostic
accuracy studies involving binormal ROC curve indices, Statistic in medicine, Vol. 16,
1529-1542, 1997.
[22] Zhou, X. H., Obuchowski, N. A., and McClish D. K., Statistical Methods in Diagnostic
Medicine, John Wiley and Sons, New York, 2002.
[23] Katzenbeisser, W., and Hackel, P., An alternative to the Kolmogorov-Smirnov two sample test. Commun. In Statist.-Theor. Math., 15, 1163-1177,1986
[24] Wu, C.Y., Ma, M.C. and Lin, J.P., Evaluation of Statistical Methods for Equivalence and Non-inferiority Based on the Area Under ROC Curves, submitted to statistics in Medicine, 2004.