Adjusted R-Square

From Displayr
Jump to: navigation, search

A modification of the R-Square statistic which takes into account the number of variables in the model (i.e., more variables translates to a lower R-Square if the extra variables do not explain a non-trivial amount of the variance of the dependent variable).

Use of the Adjusted R-Square statistic is not considered good practice when comparing models and Information Criteria are generally recommended to be used instead of the Adjusted R-Square statistic.