Partial Factor Modeling: Predictor-Dependent Shrinkage for Linear Regression
Partial Factor Modeling: Predictor-Dependent Shrinkage for Linear Regression
复制标题
DOI：
10.1080/01621459.2013.779843
复制
发表时间：
2013-09-01
影响因子：
3.7
通讯作者：
Mukherjee, Sayan
中科院分区：
文献类型：
--
作者：
Hahn, P. Richard;Carvalho, Carlos M.;Mukherjee, Sayan
我们开发了一种改进的高斯因子模型，目的是诱导线性回归的预测依赖收缩。新模型可以根据真实数据和模拟数据对各种协方差结构进行良好预测。此外，新模型有助于在相关预测变量的情况下进行变量选择，这通常会阻碍其他方法。
We develop a modified Gaussian factor model for the purpose of inducing predictor-dependent shrinkage for linear regression. The new model predicts well across a wide range of covariance structures, on real and simulated data. Furthermore, the new model facilitates variable selection in the case of correlated predictor variables, which often stymies other methods.