论文信息 - R-squared for Bayesian Regression Models

R-squared for Bayesian Regression Models

Abstract The usual definition of R2 (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.

[1] Aki Vehtari,et al. Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC , 2015, Statistics and Computing.

[2] N. Nagelkerke,et al. A note on a general definition of the coefficient of determination , 1991 .

[3] Inge S. Helland,et al. On the Interpretation and Use of R 2 in Regression Analysis , 1987 .

[4] J. Gabry,et al. Bayesian Applied Regression Modeling via Stan , 2016 .

[5] Andrew Gelman,et al. Bayesian Measures of Explained Variance and Pooling in Multilevel (Hierarchical) Models , 2006, Technometrics.

[6] D. Bates,et al. Fitting Linear Mixed-Effects Models Using lme4 , 2014, 1406.5823.

[7] John T. Kent,et al. Measures of dependence for censored survival data , 1988 .

[8] Aki Vehtari,et al. Understanding predictive information criteria for Bayesian models , 2013, Statistics and Computing.

[9] Tue Tjur,et al. Coefficients of Determination in Logistic Regression Models—A New Proposal: The Coefficient of Discrimination , 2009 .

[10] R Core Team,et al. R: A language and environment for statistical computing. , 2014 .

[11] R. Xu. Measuring explained variation in linear mixed effects models , 2003, Statistics in medicine.

[12] Mari Palta,et al. Properties of R2 statistics for logistic regression , 2006, Statistics in medicine.

[13] Patrick Royston,et al. A simulation study of predictive ability measures in a survival model I: Explained variation measures , 2012, Statistics in medicine.