R-squared for Bayesian Regression Models Article Swipe

View

Related Concepts

Variance (accounting) Statistics Mathematics One-way analysis of variance Bayesian probability Analysis of variance Law of total variance Variance inflation factor Econometrics Regression analysis Conditional variance Volatility (finance) Multicollinearity Business Accounting Autoregressive conditional heteroskedasticity

Andrew Gelman , Ben Goodrich , Jonah Gabry , Aki Vehtari ·

YOU? · · 2018 · Open Access · · DOI: https://doi.org/10.1080/00031305.2018.1549100 · OA: W2904925972

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.