Generalisation of Receiver operating characteristic (ROC) curve has become increasingly useful in evaluating theperformance of diagnostic tests that have more than binary outcomes. While parametric approaches have been widelyused over the years, the limitations associated with parametric assumptions often make it difficult to modelling thevolume under surface for data that do not meet criteria under parametric distributions. As such, estimation of ROCsurface using nonparametric approaches have been proposed to obtained insights on available data. One of the commonapproaches to non-parametric estimation is the use of Bayesian models where assumptions about priors can be madethen posterior distributions obtained which can then be used to model the data. This study uses Polya tree priors wheremixtures of Polya trees approach was used to model simulated three-way ROC data. The results of VUS estimationwhich is considered a suitable inference in evaluating performance of a diagnostic test, indicated that the mixtures ofPolya trees model fitted well the ROC surface data. Further, the model performed relatively well compared toparametric and semiparametric models under similar assumptions.
展开▼
