线性结构方程参数估计的一种简单方法
A Simple Approach for Estimating Parameters in a Linear Structural Equation Model
-
摘要: 利用极大似然法或者最小二乘法等对线性结构方程进行估计时,会受到一定的限制.针对这一弊端,本文提出利用偏最小二乘法的思想估计线性结构方程,并将它应用于顾客满意度模型中.实证分析表明了该方法的有效性.Abstract: Popularly used iterative algorithms, such as maximum likelihood estimation, is limited in estimating the linear structural equation model. We suggest a non-iterative approach motivated by partial least squares regression. This approach is applied to a real dataset from a study of customer satisfaction index model is Shanghai. The application evidence the validation and the efficiency of our propose approach.