Activity Number:
|
48
|
Type:
|
Contributed
|
Date/Time:
|
Sunday, August 11, 2002 : 4:00 PM to 5:50 PM
|
Sponsor:
|
Business & Economics Statistics Section*
|
Abstract - #301359 |
Title:
|
Testing for Duration Dependence with Discrete Data
|
Author(s):
|
James McDonald*+ and Scott Grimshaw and Grant McQueen
|
Affiliation(s):
|
Brigham Young University and Brigham Young University and Brigham Young University
|
Address:
|
, Provo, , 84602,
|
Keywords:
|
Duration Dependence ; Discrete Data ; Continuous Distributions
|
Abstract:
|
Many economic studies fit continuous probability density functions to continuous data that are reported in a grouped or discrete format. Examples are especially common in the analysis of duration dependence. Monte Carlo methods are used to examine the consequences of this type of specification error for duration dependency. The simulation results suggest that inconsistent parameter estimators are obtained with an upward bias in the likelihood ratio test. This bias gives mistaken evidence of duration dependence for sample sizes and parameter values likely to arise in empirical studies. After reporting the sample sizes and parameter values that result in a significant bias, we present and evaluate two approaches that circumvent the problem
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2002 program |