MULTINOMIAL LOGISTIC-REGRESSION ALGORITHM
MULTINOMIAL LOGISTIC-REGRESSION ALGORITHM
复制标题
DOI：
10.1007/bf00048682
复制
发表时间：
1992-03-01
影响因子：
1
通讯作者：
BOHNING, D
中科院分区：
文献类型：
--
作者：
BOHNING, D
下界原理（Bohning 和 Lindsay (1988, Ann. Inst. Statist. Math., 40, 641-663)、Bohning (1989, Biometrika, 76, 375-383) 中介绍的包括用 Loewner 排序中的全局下界替换二阶导数矩阵。该下界用于 Newton-Raphson 迭代而不是 Hessian 矩阵导致单调收敛的迭代序列。在这里，我们将这一原理应用于多项逻辑回归模型，它变得特别有吸引力。
The lower bound principle (introduced in Bohning and Lindsay (1988, Ann. Inst. Statist. Math., 40, 641-663), Bohning (1989, Biometrika, 76, 375-383) consists of replacing the second derivative matrix by a global lower bound in the Loewner ordering. This bound is used in the Newton-Raphson iteration instead of the Hessian matrix leading to a monotonically converging sequence of iterates. Here, we apply this principle to the multinomial logistic regression model, where it becomes specifically attractive.