Date on Honors Thesis

Spring 4-26-2023


Mathematics/ Data Science

Examining Committee Member

Manoj Pathak, PhD, Advisor

Examining Committee Member

Warren Edminster, PhD, Executive Director Honors College

Examining Committee Member

Donald Adongo, PhD, Committee Member

Examining Committee Member

Christopher Mecklin, PhD, Committee Member


Glaucoma is a group of eye diseases in which damage gradually occurs to the optic nerve, which often leads to partial or complete loss of vision. As the second leading cause of blindness, there is no cure for glaucoma. Early detection and the tracking of its progression is key to managing the effects of glaucoma. Ordinary Least Squares Regression (OLSR), the most commonly used methodology for tracking glaucoma progression, is inappropriate as the longitudinally collected perimetry data from the glaucoma patients appears to be temporally correlated. Time series models, that account for temporal correlation, are better methods to analyze Mean Deviation (MD) series, a global indicator of disease status in glaucoma patients. In this study, we attempt to analyze MD series using an Irregular Autoregressive (IAR) model and compare its predictability with OLSR and a regular Autoregressive (AR) model.