# deletion residuals in linear [duplicate]

Possible Duplicate:
Deletion residuals

The $i^{th}$ deletion residual $e_{−i}$ is defined as $e_{−i}=y_{i}−X^{⊤}B_{−i}$ where $X^{⊤}$ is the $i^{th}$ row of the design matrix $X$ and $B_{−i}$ is a column vector of least square parameter estimates calculated without the $i^{th}$ observation. Write some annotated R code to calculate the deletion residuals when the linear model $y_{i}=B_{0}+B_{1}X_{i}+B_{2}X_{2i}+E_{i}$ is fitted to the data in the file quadratic.txt. By drawing suitable plot, comment on the distribution of these deletion residuals.

should I write what the exactly distribution?

-

## marked as duplicate by Andy W, jbowman, whuber♦Dec 5 '12 at 1:51

Please don't copy and paste homework assignments verbatim. Put some effort into it and show what you have done so far. As is the last sentence doesn't even make sense. – Andy W Dec 4 '12 at 23:52

$$e_{(i)}=\frac{e_i}{1-h_{ii}}$$
where $e_i = y_i - \hat{y_i}$ and $h_{ii}$ is the $i^{th}$ element of the diagonal of the matrix $H = X(X^tX)^{-1}X^t$