Abstract
We consider the estimation of three-dimensional ROC surfaces for continuous tests given covariates. Three way ROC analysis is important in our motivating example where patients with Alzheimer’s disease are usually classified into three categories and should receive different category-specific medical treatment. There has been no discussion on how covariates affect the three way ROC analysis. We propose a regression framework induced from the relationship between test results and covariates. We consider several practical cases and the corresponding inference procedures. Simulations are conducted to validate our methodology. The application on the motivating example illustrates clearly the age and sex effects on the accuracy for Mini-Mental State Examination of Alzheimer’s disease.
Similar content being viewed by others
References
Dabrowska D M, Doksum K A. Partial likelihood in transformation models with censored data. Scand J Statist, 1988, 15: 1–24
Efron B, Tibshirani R J. An Intriduction to the Bootstrap. London: Chapman & Hall, 1993
Fine J P. Analysing competing risks data with transformation models. JRSS Ser B, 1999, 61: 817–830
Fine J P, Bosch J. Risk assessment via a robust probit model, with application to toxicology. J Amer Statist Assoc, 2000, 95: 375–382
Fine J P, Ying Z, Wei L J. On the linear transformation model with censored data. Biometrika, 1998, 85: 980–986
Foster A M, Tian L, Wei L J. Estimation for the Box-Cox transformation model without assuming parametric error distribution. J Amer Statist Assoc, 2001, 96: 1097–1101
Han A K. A non-parametric analysis of transformations. J Econom, 1987, 35: 191–209
Heckerling P S. Parametric three-way receiver operating characteristic surface analysis using Mathematica. Medical Decision Making, 2001, 20: 409–417
Janes H, Pepe M S. Adjusting for covariate effects on classification accuracy using the covariate-adjusted receiver operating characteristic curve. Biometrika, 2009, 96: 371–382
Jin Z, Ying Z, Wei L J. A simple resampling method by perturbing the minimand. Biometrika, 2001, 90: 341–353
Kalbfleisch J D. Likelihood methods and nonparametric tests. J Amer Statist Assoc, 1978, 73: 167–170
Koepsell T D, Chi Y Y, Zhou X H, et al. An alternative method for estimating efficacy of the AN1792 Vaccine for Alzheimer’s disease. Neurology, 2007, 69: 1868–1872
Lawless J F. Statistical Models and Methods for Lifetime Data. Hobolcen, NJ: John Wiley, 2003
Li J, Fine J P. ROC analysis for multiple classes and multiple categories and its application in microarray study. Biostatistics, 2008, 9: 566–576
Li J, Tai B C, Nott D J. Confidence interval for the bootstrap P-value and sample size calculation of the bootstrap test. J Nonparametr Stat, 2009, 21: 649–661
Li J, Zhang C M, Doksum K A, et al. Simultaneous confidence intervals for semiparametric logistic regression and confidence regions for the multi-dimensional effective dose. Statistica Sinica, 2010, 20: 637–659
Li J, Zhou X H. Nonparametric and semiparametric estimation of the three way receiver operating characteristic surface. J Statist Plann Inference, 2009, 139: 4133–4142
Mossman, D. Three-way ROCs. Medical Decision Making, 1999, 19: 78–89
Nakas C T, Yiannoutsos C T. Ordered multiple-class ROC analysis with continuous measurements. Statistics in Medicine, 2004, 23: 3437–3449
Newey W K. Efficient instrumental variables estimation of nonlinear models. Econometrica, 1990, 58: 809–837
Pepe M S. The Statistical Evaluation of Medical Tests for Classification and Prediction. Oxford: Oxford University Express, 2003
Pinheiro J C, Bates D M. Mixed-Effects Models in S and S-PLUS. New York: Springer, 2000
Robinson P M. Best nonlinear three-stage least squares estimation of certain econometric models. Econometrica, 1991, 59: 755–786
Shao J. Mathematical Statistics. New York: Sringer, 1999
Wang N, Ruppert D. Nonparametric estimation of the transformation in the transform-both-sides regression models. J Amer Statist Assoc, 1995, 90: 731–738
Wei L J. The accelerated failure time model: A useful alternative to the Cox regression model in survival analysis. Statistics in Medicine, 1992, 11: 1871–1879
Zhang Y, Li J. Combining multiple markers for multi-category classification: an ROC surface approach. Australian and New Zealand Journal of Statistics, 2011, 53: 63–78
Zhou X H, Castellucio P. Adjusting for non-ignorable verification bias in clinical studies for Alzheimer’s disease. Statistics in Medicine, 2004, 23: 221–230
Zhou X H, Obuchowski N A, McClish D K. Statistical Methods in Diagnositic Medicine. New York: Wiley, 2002
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Li, J., Zhou, X. & Fine, J.P. A regression approach to ROC surface, with applications to Alzheimer’s disease. Sci. China Math. 55, 1583–1595 (2012). https://doi.org/10.1007/s11425-012-4462-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11425-012-4462-3
Keywords
- receiver operating characteristic surface
- volume under ROC surface
- rank regression
- transformation model
- maximum likelihood estimation
- bootstrap