Can we here derive general expressions for Mean Squared Error and $R^2$?

63 Views Asked by At

Suppose we have the linear relationship

$$ y= \beta _1 X_1+\beta _2 X_2+\beta _3 X_3 + \cdots + \beta _n X_n $$

and we have an estimate of $y$ given by

$$ \hat{y}= \hat{\beta }_1 X_1+\hat{\beta }_2 X_2+\hat{\beta }_3 X_3 + \cdots + \hat{\beta }_n X_n $$

where $X_1 ... X_n$ are all standard normal distributions.

Can we derive exact formulas for the mean squared error of $\hat{y}$ and for $R^2$, and is there an exact relationship between the two?