Modelling stochastic order in the analysis of receiver operating characteristic data: Bayesian non‐parametric approaches

The evaluation of the performance of a continuous diagnostic measure is a commonly encountered task in medical research. We develop Bayesian non-parametric models that use Dirichlet process mixtures and mixtures of Polya trees for the analysis of continuous serologic data. The modelling approach differs from traditional approaches to the analysis of receiver operating characteristic curve data in that it incorporates a stochastic ordering constraint for the distributions of serologic values for the infected and non-infected populations. Biologically such a constraint is virtually always feasible because serologic values from infected individuals tend to be higher than those for non-infected individuals. The models proposed provide data-driven inferences for the infected and non-infected population distributions, and for the receiver operating characteristic curve and corresponding area under the curve. We illustrate and compare the predictive performance of the Dirichlet process mixture and mixture of Polya trees approaches by using serologic data for Johne's disease in dairy cattle. Copyright (c) 2008 Royal Statistical Society.

[1]  T. Ferguson.  A Bayesian Analysis of Some Nonparametric Problems , 1973 .

[2]  M. Pepe.  The Statistical Evaluation of Medical Tests for Classification and Prediction , 2003 .

[3]  Chris Lloyd,et al.  Using Smoothed Receiver Operating Characteristic Curves to Summarize and Compare Diagnostic Systems , 1998 .

[4]  Fernando A. Quintana,et al.  Nonparametric Bayesian data analysis , 2004 .

[5]  James E. Collins,et al.  Evaluation of Five Antibody Detection Tests for Diagnosis of Bovine Paratuberculosis , 2005, Clinical Diagnostic Laboratory Immunology.

[6]  Sonia Petrone.  Discussion on "Bayesian nonparametric inference for random distributions and related functions", by Walker, S.G., Damien, P., Laud, P.W., Smith, A.F.M. , 1999 .

[7]  D. Blackwell,et al.  Ferguson Distributions Via Polya Urn Schemes , 1973 .

[8]  M. Escobar,et al.  Bayesian Density Estimation and Inference Using Mixtures , 1995 .

[9]  W. Johnson,et al.  Modeling Regression Error With a Mixture of Polya Trees , 2002 .

[10]  J. Sethuraman.  A CONSTRUCTIVE DEFINITION OF DIRICHLET PRIORS , 1991 .

[11]  D. Bamber.  The area above the ordinal dominance graph and the area below the receiver operating characteristic graph , 1975 .

[12]  T. Hanson.  Inference for Mixtures of Finite Polya Tree Models , 2006 .

[13]  Steven N. MacEachern,et al.  Efficient MCMC Schemes for Robust Model Extensions Using Encompassing Dirichlet Process Mixture Models , 2000 .

[14]  T. Ferguson.  Prior Distributions on Spaces of Probability Measures , 1974 .

[15]  Moshe Shaked,et al.  Stochastic orders and their applications , 1994 .

[16]  A. Kottas.  Nonparametric Bayesian survival analysis using mixtures of Weibull distributions , 2006 .

[17]  Alan E. Gelfand,et al.  Nonparametric Bayesian Modeling for Stochastic Order , 2001 .

[18]  A. Gelfand,et al.  Dirichlet Process Mixed Generalized Linear Models , 1997 .

[19]  C. Antoniak.  Mixtures of Dirichlet Processes with Applications to Bayesian Nonparametric Problems , 1974 .

[20]  Adam J. Branscum,et al.  Bayesian Nonparametric Modeling and Data Analysis: An Introduction , 2005 .

[21]  M. Greiner,et al.  Principles and practical application of the receiver-operating characteristic analysis for diagnostic tests. , 2000, Preventive veterinary medicine.

[22]  M. Escobar,et al.  Markov Chain Sampling Methods for Dirichlet Process Mixture Models , 2000 .

[23]  K. Zou,et al.  Smooth non-parametric receiver operating characteristic (ROC) curves for continuous diagnostic tests. , 1997, Statistics in medicine.

[24]  W. Michael Conklin,et al.  Monte Carlo Methods in Bayesian Computation , 2001, Technometrics.

[25]  Chap T Le,et al.  A solution for the most basic optimization problem associated with an ROC curve , 2006, Statistical methods in medical research.

[26]  Alan E. Gelfand,et al.  Modeling Variability Order: A Semiparametric Bayesian Approach , 2001 .

[27]  A. Gelfand,et al.  Nonparametric Bayesian bioassay including ordered polytomous response , 1991 .

[28]  S. Geisser,et al.  A Predictive Approach to Model Selection , 1979 .

[29]  Purushottam W. Laud,et al.  Bayesian Nonparametric Inference for Random Distributions and Related Functions , 1999 .

[30]  Alan E Gelfand,et al.  A Nonparametric Bayesian Modeling Approach for Cytogenetic Dosimetry , 2002, Biometrics.

[31]  Wesley O Johnson,et al.  Bayesian semiparametric ROC curve estimation and disease diagnosis , 2008, Statistics in medicine.

[32]  M. Lavine.  More Aspects of Polya Tree Distributions for Statistical Modelling , 1992 .

[33]  Alaattin Erkanli,et al.  Bayesian semi‐parametric ROC analysis , 2006, Statistics in medicine.

[34]  Wesley O. Johnson,et al.  Bayesian inferences for receiver operating characteristic curves in the absence of a gold standard , 2006 .