一类改进的贝叶斯模型选择方法

An Improved Bayesian Model Selection Method

文章针对线性回归问题中的模型选择进行了研究,从理论上提出了将k折交叉验证与投影预测法相结合的模型选择方法,通过数据仿真验证了此方法的合理性。

This paper studies the model selection in linear regression, and theoretically puts forward the model selection method by combining k-fold cross-validation and projection prediction method. Finally the paper verifies the rationality of this method by data simulation.