Generalized Cross Entropy Method for estimating joint distribution from incomplete information
详细信息 查看全文
- 作者：Hai-Yan Xua ; xuh@ihpc.a-star.edu.sg" class="auth_mail" title="E-mail the corresponding author ; Shyh-Hao Kuoa ; Guoqi Lia ; b ; Erika Fille T. Legaraa ; Daxuan Zhaoa ; c ; Christopher P. Monterolaa ; monterolac@ihpc.a-star.edu.sg" class="auth_mail" title="E-mail the corresponding author
- 关键词：Maximum entropy ; Minimum discrimination information ; Joint distribution ; KL distance ; Demography ; Household profile
- 刊名：Physica A: Statistical Mechanics and its Applications
- 出版年：2016
- 出版时间：1 July 2016
- 年：2016
- 卷：453
- 期：Complete
- 页码：162-172
- 全文大小：609 K
文摘
- •
A new and novel algorithm named Generalized Cross Entropy Model (GCEM) is proposed.
- •
GCEM estimates full joint distribution from marginal even with incomplete information.
- •
Existing maximum entropy procedures are shown to be just special cases of GCEM.
- •
Accuracy of GCEM is established and illustrated using actual empirical data.
- •
Article provides guide on how GCEM could be applied to diverse fields/areas.