https://scholars.lib.ntu.edu.tw/handle/123456789/428481
標題: | Bias and variance reduction in nonparametric estimation of time-dependent accuracy measures | 作者: | CHIN-TSANG CHIANG Huang, M.-Y. Wang, S.-H. |
關鍵字: | bandwidth selection; conditional survival function; Gaussian process; kernel function; marker-dependent censoring; positive/negative predictive value; receiver operating characteristic curve; true/false positive rate; U-statistic | 公開日期: | 2016 | 卷: | 35 | 期: | 28 | 起(迄)頁: | 5247-5266 | 來源出版物: | Statistics in Medicine | 摘要: | A new nonparametric approach is developed to estimate the time-dependent accuracy measure curves, which are defined on the cumulative cases and dynamic controls, for censored survival data. Based on an estimable survival process, the main intention of this study is to reduce the finite-sample biases of nearest neighbor estimators. The asymptotic variances of some retrospective accuracy measure estimators are further reduced by applying a smoothing technique to the underlying process of a marker. Meanwhile, practically feasible and theoretically valid procedures are proposed for bandwidth selection in the presented estimators. In addition, the proposed methodology can be reasonably extended to accommodate stratified survival data and survival data with multiple markers. As shown in the simulations, our new estimators outperform the nearest neighbor and inverse censoring weighted estimators. Data from the AIDS Clinical Trials Group study 175 and an angiographic coronary artery disease study are also used to illustrate the proposed methodology. Copyright ? 2016 John Wiley & Sons, Ltd. Copyright ? 2016 John Wiley & Sons, Ltd. |
URI: | https://scholars.lib.ntu.edu.tw/handle/123456789/428481 | DOI: | 10.1002/sim.7058 | SDG/關鍵字: | accuracy; Article; bootstrapping; human; Human immunodeficiency virus infection; k nearest neighbor; kernel method; mathematical analysis; mathematical model; statistical bias; variance; computer simulation; coronary artery disease; reproducibility; retrospective study; biological marker; Bias; Biomarkers; Computer Simulation; Coronary Artery Disease; Humans; Reproducibility of Results; Retrospective Studies |
顯示於: | 應用數學科學研究所 |
在 IR 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。