Advances and Applications in Statistics
Volume 49, Issue 4, Pages 305 - 326
(October 2016) http://dx.doi.org/10.17654/AS049040305 |
|
A GAMMA REGRESSION FOR BOUNDED CONTINUOUS VARIABLES
Amany M. Mousa, Ahmed A. El-Sheikh and Mahmoud A. Abdel-Fattah
|
Abstract: This paper presents a regression model for continuous variables taking values in a bounded interval based on the unit gamma distribution. The model may serve as an alternative to the beta regression model. The score vector and Fisher’s information matrix are derived. Two real applications are included. A simulation study is carried out to compare the unit gamma regression model with the beta regression model in terms of different goodness of fit and information criteria. |
Keywords and phrases: beta regression, bounded continuous data, ordinary least squares, unit gamma regression. |
|
Number of Downloads: 590 | Number of Views: 2729 |
|