2

I am assuming the following model:

$Y = \beta X + \epsilon$

Here both $X$ and $Y$ are matrices. I fit the least squares model without any regularization and get the matrix $\beta$. I would like to do the following now. I want to pick some subset $S$ of rows of $\beta$ using forward stepwise selection such that these rows minimize the squared loss. How should one go about it?

sashkello
  • 2,198
  • 1
  • 20
  • 26
Shishir Pandey
  • 1,051
  • 2
  • 9
  • 11
  • 7
    It's probably best to *avoid* going about it. If that doesn't make sense / you want to understand why, it may help to read my answer here: [algorithms-for-automatic-model-selection](http://stats.stackexchange.com/questions/20836//20856#20856). Why are you opposed to regularization? – gung - Reinstate Monica Apr 09 '13 at 04:20
  • I am not opposed to regularization. Its just that I was not planning to use it for the case now. – Shishir Pandey Apr 11 '13 at 10:08

0 Answers0