|
Activity Number:
|
506
- New Ideas in Inference
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 6, 2020 : 10:00 AM to 2:00 PM
|
|
Sponsor:
|
International Indian Statistical Association
|
|
Abstract #314067
|
|
|
Title:
|
Multivariate Regression Model for Doubly Inflated Counts That Employs Gaussian Copula
|
|
Author(s):
|
Monika Arora* and N. Rao Chaganty
|
|
Companies:
|
Indraprastha Institute of Information Technology, Delhi and Old Dominion University
|
|
Keywords:
|
|
|
Abstract:
|
Bivariate count data are commonly modeled using bivariate Poisson distributions. The zero inflation in bivariate count data was studied by Lee et al. (2009) and further extended by Sengupta et al. (2016) to handle doubly inflated counts. There could be scenarios where the existing bivariate Poisson models are not able to capture the actual dispersion. For such data Sellers et al. (2016) have used bivariate Conway-Maxwell-Poisson (CMP) distribution. An alternative method to construct a bivariate CMP distribution uses Gaussian copula with Conway-Maxwell-Poisson marginals. In this paper, we propose a general copula-based multivariate regression model with univariate CMP mixture marginals to handle correlated and doubly inflated counts. The proposed model has various applications in research areas like healthcare and biology.
|
Authors who are presenting talks have a * after their name.
