Adjusted R2

CFA Level II Quant
#1

Hi, suppose the following regression output. It’s 1 independent variable (simple regression):
image

I get the idea of “R square”, basically our regression model is explaining 40% of the variation in Y.

What does the 25% in “Adjusted R square” means? I know its calculation and that it comes to penalize the fact that sometimes adding an independent variable doesn’t add explanatory power, but actually not quite sure how to translate into words the meaning of this 25%. Any insight will be appreciated.

#2

By itself, it means pretty much nothing.

Its usefulness depends on comparing two regressions: one with an additional independent variable missing in the other. You compare the adjusted R2 values and select the regression with the higher value.

In short; it’s a relative measure, not an absolute measure.

This might help a bit.

1 Like