利用统计曲率的概念，研究结构方程模型的最大似然估计量和广义最小二乘估计量的信息损失，得到了简明的结果。
By employing the concept of statistical curvatures, the information loss of the maximum likelihood estimator and the generalized least squares estimator is investigated.
利用广义逆矩阵的理论，本文给出了完整的最小二乘递推算法，并由此得到若干很有意义的推论。
Using the theory of generalized inverse matrix, a perfect algorithm of recurrence for least square has been derived. And from this algorithm, some corollaries of great significance were obtained.
讨论了线性流形上广义次对称矩阵反问题的最小二乘解及其逼近问题。
The least squares solution of inverse problems of generalized skew symmetric matrices and It's optimal approximation problems are discussed.
而广义移动最小二乘近似要求近似函数及其导数在所有节点处的误差的平方和最小。
However, the generalized moving least squares approximation makes require least squares approximation with regard to functional and its derivative value on all nodes.
在误差为相依的情况下，讨论了线性回归模型的刀切最小二乘估计与广义刀切最小二乘估计。
This paper studies linear regression models with dependent errors, and we introduce the jackknifed least squares estimator and generalized jackknife least squares estimator.
结果显示广义应变花法应变的计算精度比最小二乘配置法高。
The result shows that GSRM has higher accuracy than LSCM in strain calculation.
本文通过建立节点位移和广义位移之间的关系对移动最小二乘形函数进行修正，给出了修正的移动最小二乘形函数；
The modified MLS shape function, for alleviating the above problem, is given by establishing the relationship between the nodal value and the generalized displacement.
即利用矩阵分析理论中的广义逆法，完成非线性方程组的最小二乘求解过程。
The paper also proposes the relevant numerical solver. Namely, it has Solved the equations uses the generalized reciprocal method in the theory of matrix analysis.
比较分析了最小二乘支持向量机（LSSVM）和广义回归神经网络（GRNN）这两种方法的特点。
The features of two methods, i. e. least square support vector machine (LSSVM) and generalized regression neural network (GRNN) are compared and analyzed.
结果显示广义应变花法应变的计算精度比最小二乘配置法高。
The result shows that GSRM has higher accuracy than LSCM in strain...
结果显示广义应变花法应变的计算精度比最小二乘配置法高。
The result shows that GSRM has higher accuracy than LSCM in strain...
应用推荐