ESTIMATION IN LINEAR-MODELS USING GRADIENT DESCENT WITH EARLY STOPPING
ESTIMATION IN LINEAR-MODELS USING GRADIENT DESCENT WITH EARLY STOPPING
复制标题
DOI：
10.1007/bf00156750
复制
发表时间：
1994-12-01
影响因子：
2.2
通讯作者：
BRAMSON, MJ
中科院分区：
文献类型：
--
作者：
SKOURAS, K;GOUTIS, C;BRAMSON, MJ
本文导出了线性模型系数的一种新的收缩估计。 该估计器的动机是梯度下降算法，用于最小化平方误差和结果的总和，从早期停止的算法。 的估计的统计特性进行检查，并与其他行之有效的方法，如最小二乘法和岭回归，无论是分析和通过模拟研究进行比较。 一个重要的结果是，新的估计被证明是可比其他收缩估计的参数和预测的均方误差，并在某些情况下，上级。
A new shrinkage estimator of the coefficients of a linear model is derived. The estimator is motivated by the gradient-descent algorithm used to minimize the sum of squared errors and results from early stopping of the algorithm. The statistical properties of the estimator are examined and compared with other well-established methods such as least squares and ridge regression, both analytically and through a simulation study. An important result is that the new estimator is shown to be comparable to other shrinkage estimators in terms of mean squared error of parameters and of predictions, and superior under certain circumstances.