Abstract

The Rayleigh Mixture ROC Curve can be applied when the lifetime data is heterogeneous or population consists of many subpopulations. We have proposed Rayleigh Mixture Receiver Operating Characteristic (ROC) model and also discussed its properties. Area under the Rayleigh Mixture ROC Curve (AUC) and its Variance are also found. Estimates of the parameters of Rayleigh Mixture ROC Curve are derived by using the Maximum Likelihood Method (EM) algorithm and Method of Moments. The proposed model is validated by using the simulation studies and real life example of head trauma data. It is found that the estimates using EM algorithm are closer to parameters as compared to the estimates using Method of Moments for large sample sizes.

Author: Sudesh Pundir and Azhar Uddin

Received on: July, 2019

Accepted on: July, 2020