Whiles reading An Introduction To Statistical Learning under linear regression (Chapter 3), I found:

$$E(Y - \hat{Y})^2 = |f(X) - \hat{f}(X)|^2 + Var(ε)$$

where $E(Y - \hat{Y})^2$ represents the average, or expected value of the squared difference between the predicted and actual value of $Y$, and $Var(ε)$ represents the variance associated with the error term $ε$.

What does $Var(ε)$ mean? Can we just write $ε$ rather than $Var(ε)$?

  • $\begingroup$ can you share the author and the page? and edition if applicable. $\endgroup$
    – gunes
    Feb 16 '19 at 18:49
  • $\begingroup$ @Gunes see stats.stackexchange.com/…. $\endgroup$
    – whuber
    Feb 16 '19 at 20:03
  • $\begingroup$ Now I found it, thanks. It's in Page 19, not in Chapter 3. I was interested in finding some evidence about $X$ being fixed; and it is there. However, it appears the question is duplicate. $\endgroup$
    – gunes
    Feb 16 '19 at 20:11