Colloquium Speaker:
Hossein Varzdar (PhD Candidate, Statistics, Univerity of Windsor)
Title: The Transformed Bivariate Generalized Exponential Distribution via Clayton Copula: Estimation and Regression
Abstract: Bivariate Generalized Exponential distributions have recently proven effective for modeling diverse forms of bivariate data. This paper introduces a new class of bivariate distribution constructed via a copula-based through transformation of the Bivariate Generalized Exponential distribution using the Clayton copula originally introduced by Kundu, Kumar, and Gupta (2015). Two estimation procedures are proposed: a Maximization-by-Parts (MBP) method and a two-stage copula-based approach. We also develop regression models for correlated proportions on (0,1)×(0,1), and establish the asymptotic properties of the resulting estimators under mild regularity conditions. Simulation studies show that both estimation methods produce consistent and comparable results, particularly in large samples. A real-data application further demonstrates the practical utility and flexibility of the proposed framework.
Day & Time: Thursday, November 27, 2025, at 3:00pm
Location: Lambton Tower, Room 9-118