R-squared for Bayesian Regression Models
R-squared for Bayesian Regression Models
复制标题
DOI：
10.1080/00031305.2018.1549100
复制
发表时间：
2019-07-03
影响因子：
1.8
通讯作者：
Vehtari, Aki
中科院分区：
文献类型：
--
作者：
Gelman, Andrew;Goodrich, Ben;Vehtari, Aki
R-2的通常定义（预测值的方差除以数据的方差）对于贝叶斯拟合的问题存在问题，因为分子可以大于分母。我们提出了一个类似于生存分析文献中出现的替代定义：预测值的差异除以预测值的方差加上误差的预期差异。
The usual definition of R-2 (variance of the predicted values divided by the variance of the data) has a problem for Bayesian fits, as the numerator can be larger than the denominator. We propose an alternative definition similar to one that has appeared in the survival analysis literature: the variance of the predicted values divided by the variance of predicted values plus the expected variance of the errors.