Download pdf
Comparison of methods for calculating confidence intervals of AUC in ROC curve considering sampling error
- 1 Peking University
- 2 Southeast University
- 3 University of Cambridge
* Author to whom correspondence should be addressed.
Abstract
The Receiver Operating Characteristic (ROC) curve is a crucial method for evaluating the effectiveness of diagnostic medical indicators and has found extensive applications. However, errors are inevitable in the data acquisition process. Therefore, discussions on error and various methods for improving and handling data have not only become the focus of academic discourse but also hold practical significance. Unlike general statistics, the diversity of error situations, ranges, and impacts in biostatistics often present unique challenges. In practical scenarios, such as drug experiments, limited sample sizes and variations in individual responses to the same drug necessitate the use of error models, data scales, and statistical processing based on historical data, biomedical knowledge, and experimental data. Furthermore, the choice of an appropriate method depends on the specific objectives of the experiment, which is essential for producing compelling conclusions. Importantly, the field of biology has introduced methods to address errors, such as cross-comparison experiments or repeated experiments, and data processing must adapt to changes in experimental designs. This paper presents a statistical approach based on the widely used practice of error reduction through repeated experiments in the context of assessing generic drug consistency. The paper first summarizes the common types of errors encountered in biostatistics and the corresponding analytical, control, and optimization measures. It explores several methods for calculating the Area Under the ROC Curve (AUC) when sampling error is introduced and applies error reduction through repeated experiments. Subsequently, the paper validates the methods under different error scenarios using simulated data, highlighting the suitability of different statistical models and their reasons for selection in cases where the difference between healthy and diseased populations is not substantial. This paper offers valuable insights into handling various types of real-world data to eliminate errors and obtain more accurate statistical conclusions.
Keywords
ROC curve, AUC, confidence interval, sampling error
[1]. Zou,K.H, Hall,W.J. and Shapiro,D.E.‘Smooth non-parametric receiver operating characteristic(ROC) curves for continuous diagnostic tests’, Statistics in Medicine, 16, 2143-2156 (1997).
[2]. Wieand, S., Gail M. H., James, B.R. and James K. L.‘A family of non-parametric statistics for comparing diagnostic markers with paired or unpaired data’,Biometrika, 76, 585-592 (1989).
[3]. Bamber, D.C. ‘The area above the ordinal dominance graph and the area below the receiver operating characteristic graph’,Journal of Mathematical Psychology, 12, 387-415 (1975).
[4]. David F.‘The effect of random measurment error on receiver operating characteristic (ROC) curves’,Statictics in Medicine, 19, 61-70 (2000).
[5]. Owen,D.B., Craswell,K.J. and Hanson,D.L.‘Non-parametric upper confidence bound for P(Y<X) and confidence limits for P(Y<X) when X and Y are normal’Journal of the American Statistical Association, 59, 906-924 (1964).
[6]. Thomas J.D., Hultquist R.A. ‘Interval estimation for the unbalanced case of the one-way random effect model’, Annals of Statistics, 6, 582-587 (1978)
[7]. Yanhong L., John J.K., Allan D. and Zou G.Y.‘Interval estimation for the area under the receiver operating characteristic curve when data are subject to error’ Statistics in Medicine, 29, 2521-2531 (2010)
[8]. Howe WG. ‘Approximate confidence limits on the mean of X+Y where X and Y are two tabled independent random variable’,Journal of the American Statistical Association, 69, 789-794 (1974)
[9]. Graybill F.A. and Wang C.M. ‘Confidence intervals on nonnegative linear combination of variances’, Journal of the American Statistics Association. 75, 869-873 (1980)
Cite this article
She,Y.;Cui,J.;Liu,X. (2024). Comparison of methods for calculating confidence intervals of AUC in ROC curve considering sampling error. Applied and Computational Engineering,46,189-198.
Data availability
The datasets used and/or analyzed during the current study will be available from the authors upon reasonable request.
Disclaimer/Publisher's Note
The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of EWA Publishing and/or the editor(s). EWA Publishing and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
About volume
Volume title: Proceedings of the 4th International Conference on Signal Processing and Machine Learning
© 2024 by the author(s). Licensee EWA Publishing, Oxford, UK. This article is an open access article distributed under the terms and
conditions of the Creative Commons Attribution (CC BY) license. Authors who
publish this series agree to the following terms:
1. Authors retain copyright and grant the series right of first publication with the work simultaneously licensed under a Creative Commons
Attribution License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this
series.
2. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the series's published
version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial
publication in this series.
3. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and
during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See
Open access policy for details).