smoothed empirical likelihood inference for roc curves with missing dataroc曲线的平滑的经验似然推断缺失的数据.pdfVIP

Open Journal of Statistics, 2012, 2, 21-27 /10.4236/ojs.2012.21003 Published Online January 2012 (http://www.SciRP.org/journal/ojs) Smoothed Empirical Likelihood Inference for ROC Curves with Missing Data Yueheng An Department of Mathematics and Statistics, Georgia State University, Atlanta, USA Email: yan3@ Received September 25, 2011; revised November 4, 2011; accepted November 16, 2011 ABSTRACT The receiver operating characteristic (ROC) curve has been widely used in scientific research fields. After using the random hot deck imputation, we propose the smoothed empirical likelihood ratio statistic for the ROC curve with miss- ing data. Its asymptotic distribution is a scaled chi-square distribution and empirical likelihood confidence intervals for ROC curves are constructed. The simulation study shows that the proposed interval estimates perform well based on the coverage probability for different sample sizes and response rates. Keywords: Confidence Interval; Missing Data; ROC Curve; Smoothed Empirical Likelihood 1. Introduction Qin and Qian [10] proposed smoothing EL interval esti- mation for the difference of two quantiles with missing The receiver operating characteristic (ROC) curve has data. Motivated by their idea, we propose empirical like- been extensively used to evaluate the diagnostic tests. lihood ratio for the ROC curve with missing data a