4.6 Article

On regularization of generalized maximum entropy for linear models

期刊

SOFT COMPUTING
卷 25, 期 12, 页码 7867-7875

出版社

SPRINGER
DOI: 10.1007/s00500-021-05805-2

关键词

Entropy; Generalized maximum entropy; Lasso regression; Regularization; Regularized generalized maximum entropy; Ridge regression

资金

  1. Centre of Excellence in Econometrics, Chiang Mai University

向作者/读者索取更多资源

Motivated by advancements in parameter estimation in linear models using regularization methods like Ridge and Lasso, the study focuses on regularization in Generalized Maximum Entropy, showing its superior performance and potential for further theoretical research. An application of the method is demonstrated with data from Thailand on the impact of education on economic growth.
Motivated by the advances in the estimation of parameters in linear models by regularization methods such as Ridge and Lasso regularizations, we investigate regularization of Generalized Maximum Entropy, which is an alternative estimation method in linear models. Our simulations confirm the better performance of the regularized Generalized Maximum Entropy estimation method, which could stimulate further theoretical research. An application of the new estimation method is illustrated with data from Thailand concerning the effect of education on economic growth.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.6
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据