Activity Number:
|
658
- Biometrics Data Mining
|
Type:
|
Contributed
|
Date/Time:
|
Thursday, August 3, 2017 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Biometrics Section
|
Abstract #324815
|
|
Title:
|
A COM-Type Generalization of the Negative Binomial Distribution with Applications
|
Author(s):
|
Ram Tripathi* and Sudip Roy and Narayanaswamy Balakrishnan
|
Companies:
|
UTSA, Texas and University of Texas at San Antonio and McMaster University
|
Keywords:
|
Over-dispersion ;
MLE ;
log-concavity ;
failure rate ;
shape parameter ;
likelihood ratio test
|
Abstract:
|
Negative Binomial (NB) distribution is a widely used discrete probability distribution that arises when drawing sample with replacement from an infinite population. This distribution is always over-dispersed, which we considered as a limitation. We propose a modified version of the NB called the COM-Negative Binomial distribution (COM-NB) by introducing a shape parameter. We note that recently, Chakraborty and Ong proposed another COM-NB model with a different approach and orientation. COM-NB can display lower and higher index of dispersion than its ordinary counterpart. COM-NB approaches to COM-Poisson distribution under suitable limiting conditions. Some salient characteristics of the model such as moments, log-concavity and log-convexity and monotonicity of the failure rates are investigated. We have developed MLE of the parameters of the COM-NB and examined the behavior extensively by simulation. We have formulated likelihood ratio test for the shape parameter to ascertain if the more general model is appropriate for a given data set. We present some examples on its applicability in various areas by fitting the model to some available data from literature.
|
Authors who are presenting talks have a * after their name.