|
Activity Number:
|
108
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
IMS
|
| Abstract - #310410 |
|
Title:
|
Using Copula To Study Dependence in Mixed Distributions
|
|
Author(s):
|
Magdalena Niewiadomska-Bugaj*+
|
|
Companies:
|
Western Michigan University
|
|
Address:
|
Everett Tower, Kalamazoo, MI, 49008,
|
|
Keywords:
|
zero inflated data ; mixed distributions ; dependence ; copula
|
|
Abstract:
|
Nonnegative continuous distributions with a positive probability at zero are often used for modeling in the medical, environmental and other research areas. Most of the existing statistical techniques cannot be used for analysis of such data because the necessary assumptions (e.g., continuity, normality) are violated. This paper is devoted to studying dependence in nonnegative variables with a positive probability at zero. Through the grade transformation the mixed distribution becomes a copula and the dependence can then be studied accordingly. We will focus on implications, advantages and limitations of the proposed approach.
|
