Conference Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 115 - Advances in Clustering and Classification
Type: Contributed
Date/Time: Monday, August 8, 2022 : 8:30 AM to 10:20 AM
Sponsor: Section on Statistical Learning and Data Science
Abstract #322461
Title: Application of Machine Learning for Predicting Outcomes in a Random Effect Clustered Bivariate Model
Author(s): Edmund Essah Ameyaw* and Seth Akonor Adjei and John Kwagyan
Companies: Howard University and Northern Kentucky University and Howard University
Keywords: Machine Learning; Predictive Models; Random Effect Models; Clustered Bivariate Outcomes

Random effect model is routinely used in the analyses of clustered data. However, the use of clustered bivariate models for prediction is very limited. We report the development and application of a random effect (logistic-Gaussian) model for clustered bivariate binary outcomes. We use a Gauss Hermite Quadrature approach for the approximation of the marginal likelihood function. We then use a series of machine learning approaches to investigate the predictive performance of the model. We apply the approach to predict vision loss data due to diabetic retinopathy patients, viewing the pair of eyes as a bivariate outcome and discuss it performance using measures of sensitivity and specificity, and report the area under the Receiver Operator Curve (ROC) curve (AUC).

Authors who are presenting talks have a * after their name.

Back to the full JSM 2022 program