| 引用本文: | 刘燕,陈英武.结构方程模型参数估计的GME方法.[J].国防科技大学学报,2007,29(1):116-121.[点击复制] |
| LIU Yan,CHEN Yingwu.Generalized Maximum Entropy Method for Estimating Parameters of Structural Equation Model[J].Journal of National University of Defense Technology,2007,29(1):116-121[点击复制] |
|
| |
|
|
| 本文已被：浏览 6998次 下载 7307次 |
| 结构方程模型参数估计的GME方法 |
| 刘燕, 陈英武 |
|
(国防科技大学 信息系统与管理学院，湖南 长沙 410073)
|
| 摘要: |
| 提出应用广义最大熵的方法对结构方程模型的参数进行估计，该方法能够处理有限的或者不完全数据。为了验证广义最大熵方法在估计结构方程模型参数时的性能，采用了一个应用于美国顾客满意指数（ACSI）的结构方程模型生成仿真数据，仿真结果表明，与PLS相比，GME方法具有更高的拟合精度和较低的预测误差，在样本数据较少时，表现更为明显。 |
| 关键词: 结构方程模型 参数估计 广义最大熵 |
| DOI： |
| 投稿日期：2006-08-30 |
| 基金项目: |
|
| Generalized Maximum Entropy Method for Estimating Parameters of Structural Equation Model |
| LIU Yan, CHEN Yingwu |
|
(College of Information System and Management, National Univ. of Defense Technology, Changsha 410073, China)
|
| Abstract: |
| IA new method based on generalized maximum entropy is applied in estimating parameters of structural equation model, which is meant for situations with limited or incomplete data. A sample that is used for ACSI is used to generate simulation data in order to verify the performance of GME. The results show that GME method can achieve higher imitating precision and smaller prediction errors compared to PLS, which is prominent when the sample size is relatively small. |
| Keywords: structural equation model parameters estimation generalized maximum entropy |
|
|
