1
$\begingroup$

I am little confused what actually is the J in the formula of the SST and SSR for multiple regression

SST= $Y^T\left[ 1-\frac{1}{n}J\right]Y$

SSR=$Y^T\left[ H-\frac{1}{n}J\right]Y$

$\endgroup$
0
$\begingroup$

$J$ is the matrix of all $1$s. i.e., let $$ \mathbf{1}=(1,1,...,1)^T\in \mathbb{R}^n, $$ then $$ J = \mathbf{1}\mathbf{1}^T. $$ While $\frac{1}{n}J$ can be called "means generating matrix", namely, for some $y=(y_1, y_2,...,y_n)^T \in \mathbb{R}^n$, then $$ \frac{1}{n}Jy= (\bar{y}_n, \bar{y}_n, ..., \bar{y}_n)^T, $$ which is an essential part of any sum of squares.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.