The Annals of Applied Statistics

Functional covariate-adjusted partial area under the specificity-ROC curve with an application to metabolic syndrome diagnosis

Vanda Inácio de Carvalho, Miguel de Carvalho, Todd A. Alonzo, and Wenceslao González-Manteiga

Full-text: Access denied (no subscription detected)

We're sorry, but we are unable to provide you with the full text of this article because we are not able to identify you as a subscriber. If you have a personal subscription to this journal, then please login. If you are already logged in, then you may need to update your profile to register your subscription. Read more about accessing full-text


Due to recent advances in technology, medical diagnosis data are becoming increasingly complex and, nowadays, applications where measurements are curves or images are ubiquitous. Motivated by the need of modeling a functional covariate on a metabolic syndrome case study, we develop a nonparametric functional regression model for the area under the specificity receiver operating characteristic curve. This partial area is a meaningful summary measure of diagnostic accuracy for cases in which misdiagnosis of diseased subjects may lead to serious clinical consequences, and hence it is critical to maintain a high sensitivity. Its normalized value can be interpreted as the average specificity over the interval of sensitivities considered, thus summarizing the trade-off between sensitivity and specificity. Our methods are motivated by, and applied to, a metabolic syndrome study that investigates how restricting the sensitivity of the gamma-glutamyl-transferase, a metabolic syndrome marker, to certain clinical meaningful values, affects its corresponding specificity and how it might change for different curves of arterial oxygen saturation. Application of our methods suggests that oxygen saturation is key to gamma-glutamyl transferase’s performance and that some of the different intervals of sensitivities considered offer a good trade-off between sensitivity and specificity. The simulation study shows that the estimator associated with our model is able to recover successfully the true overall shape of the functional covariate-adjusted partial area under the curve in different complex scenarios.

Article information

Ann. Appl. Stat., Volume 10, Number 3 (2016), 1472-1495.

Received: November 2014
Revised: April 2016
First available in Project Euclid: 28 September 2016

Permanent link to this document

Digital Object Identifier

Mathematical Reviews number (MathSciNet)

Zentralblatt MATH identifier

Arterial oxygen saturation average specificity biomarker functional covariate-adjustment gamma-glutamyl transferase kernel regression metabolic syndrome partial area under the curve sensitivity specificity-receiver operating characteristic curve


Inácio de Carvalho, Vanda; de Carvalho, Miguel; Alonzo, Todd A.; González-Manteiga, Wenceslao. Functional covariate-adjusted partial area under the specificity-ROC curve with an application to metabolic syndrome diagnosis. Ann. Appl. Stat. 10 (2016), no. 3, 1472--1495. doi:10.1214/16-AOAS943.

Export citation


  • Adimari, G. and Chiogna, M. (2012). Jackknife empirical likelihood based confidence intervals for partial areas under ROC curves. Statist. Sinica 22 1457–1477.
  • Aneiros-Pérez, G. and Vieu, P. (2006). Semi-functional partial linear regression. Statist. Probab. Lett. 76 1102–1110.
  • Cai, T. and Dodd, L. E. (2008). Regression analysis for the partial area under the ROC curve. Statist. Sinica 18 817–836.
  • Davison, A. C. and Hinkley, D. V. (1997). Bootstrap Methods and Their Application. Cambridge Series in Statistical and Probabilistic Mathematics 1. Cambridge Univ. Press, Cambridge.
  • Dodd, L. E. and Pepe, M. S. (2003). Partial AUC estimation and regression. Biometrics 59 614–623.
  • Eckel, R. H., Grundy, S. M. and Zimmet, P. Z. (2005). The metabolic syndrome. Lancet 365 1415–1428.
  • Febrero-Bande, M. and Oviedo de la Fuente, M. (2012). Statistical computing in functional data analysis: The $\mathbb{R}$ package fda.usc. J. Stat. Softw. 51 1–28.
  • Ferraty, F., Van Keilegom, I. and Vieu, P. (2010). On the validity of the bootstrap in non-parametric functional regression. Scand. J. Stat. 37 286–306.
  • Ferraty, F. and Vieu, P. (2002). The functional nonparametric model and application to spectrometric data. Comput. Statist. 17 545–564.
  • Ferraty, F. and Vieu, P. (2006). Nonparametric Functional Data Analysis: Theory and Practice. Springer, New York.
  • Gigliarano, C., Figini, S. and Muliere, P. (2014). Making classifier performance comparisons when ROC curves intersect. Comput. Statist. Data Anal. 77 300–312.
  • González-Manteiga, W., Pardo-Fernández, J. C. and Van Keilegom, I. (2011). ROC curves in non-parametric location-scale regression models. Scand. J. Stat. 38 169–184.
  • Gude, F., Rey-Garcia, J., Fernandez-Merino, C., Meijide, L., García-Ortiz, L., Zamarron, C. and Gonzalez-Quintela, A. (2009). Serum levels of gamma-glutamyl transferase are associated with markers of nocturnal hypoxemia in general adult population. Clin. Chim. Acta 407 67–71.
  • Härdle, W. (1991). Smoothing Techniques: With Implementation in S. Springer, New York.
  • Härdle, W. and Marron, J. S. (1991). Bootstrap simultaneous error bars for nonparametric regression. Ann. Statist. 19 778–796.
  • Hung, H. and Chiang, C.-T. (2011). Nonparametric methodology for the time-dependent partial area under the ROC curve. J. Statist. Plann. Inference 141 3829–3838.
  • Inácio, V., González-Manteiga, W., Febrero-Bande, M., Gude, F., Alonzo, T. A. and Cadarso-Suárez, C. (2012). Extending induced ROC methodology to the functional context. Biostat. 13 594–608.
  • Inácio de Carvalho, V., Jara, A., Hanson, T. E. and de Carvalho, M. (2013). Bayesian nonparametric ROC regression modeling. Bayesian Anal. 8 623–645.
  • Inácio de Carvalho, V., de Carvalho, M., Alonzo, T. A. and González-Manteiga, W. (2016). Supplement to “Functional covariate-adjusted partial area under the specificity-ROC curve with an application to metabolic syndrome diagnosis.” DOI:10.1214/16-AOAS943SUPP.
  • Jiang, Y., Metz, C. E. and Nishikawa, R. M. (1996). A receiver operating characteristic partial area index for highly sensitive diagnostic tests. Radiology 201 745–750.
  • Lee, D. S., Ewans, J. C., Robins, S. J., Wilson, P. W., Albano, I., Fox, C. S., Wang, T. J., Benjamin, E. J. and Vasan, R. S. (2007). Gamma glutamyl transferase and metabolic syndrome, cardiovascular disease, and mortality risk: The framingham heart study. Arteriosclorosis, Trombosis, and Vascular Biology 27 127–133.
  • López-Pintado, S. and Romo, J. (2009). On the concept of depth for functional data. J. Amer. Statist. Assoc. 104 718–734.
  • Ma, H., Bandos, A. I., Rockette, H. E. and Gur, D. (2013). On use of partial area under the ROC curve for evaluation of diagnostic performance. Stat. Med. 32 3449–3458.
  • Pardo-Fernández, J. C., Rodríguez-Álvarez, M. X. and Van Keilegom, I. (2014). A review on ROC curves in the presence of covariates. REVSTAT 12 21–41.
  • $\mathbb{R}$ Development Core Team (2011). $\mathbb{R}$: A Language and Environment for Statistical Computing. $\mathbb{R}$ Foundation for Statistical Computing, Vienna.
  • Sun, Y. and Genton, M. G. (2011). Functional boxplots. J. Comput. Graph. Statist. 20 316–334.
  • van der Vaart, A. W. (1998). Asymptotic Statistics. Cambridge Series in Statistical and Probabilistic Mathematics 3. Cambridge Univ. Press, Cambridge.
  • Wang, Z. and Chang, Y.-C. I. (2011). Marker selection via maximizing the partial area under the ROC curve of linear risk scores. Biostat. 12 369–385.
  • Yao, F., Craiu, R. V. and Reiser, B. (2010). Nonparametric covariate adjustment for receiver operating characteristic curves. Canad. J. Statist. 38 27–46.

Supplemental materials

  • Supplement to “Functional covariate-adjusted partial area under the specificity-ROC curve with an application to metabolic syndrome diagnosis”. Technical details and supplementary empirical reports. The supplement consists of three parts. The first part provides auxiliary results on the construction of our estimator. The second contains supplemental empirical analysis of the metabolic syndrome data and a comparison with simpler approaches. Finally, the third part contains an additional simulation study and R code to implement our methods.