Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: _regress (stata9)


From   "R.E. De Hoyos" <redeho@hotmail.com>
To   <statalist@hsphsun2.harvard.edu>
Subject   Re: st: _regress (stata9)
Date   Tue, 19 Feb 2008 13:14:47 -0500

Dear Zurab and Austin, thanks for your suggestions.

Rafa

----- Original Message ----- From: "Austin Nichols" <austinnichols@gmail.com>
To: <statalist@hsphsun2.harvard.edu>
Sent: Friday, February 15, 2008 5:43 PM
Subject: Re: st: _regress (stata9)



"R.E. De Hoyos" <redeho@hotmail.com>:
Also see
help mf_cross

On Feb 15, 2008 5:18 PM, Zurab Sajaia <zsajaia@hotmail.com> wrote:
Hi Rafa,

try -matrix accum- (and related commands)

Zurab


----- Original Message -----
From: "R.E. De Hoyos" <redeho@hotmail.com>
To: <statalist@hsphsun2.harvard.edu>

Sent: Friday, February 15, 2008 5:14 PM
Subject: Re: st: _regress (stata9)


> Nick, Statalisters,
>
> Define X as a NxK matrix (N=observations and K=independent variables). > I
> simply want to get X'X, where N is a very large number and K is > relatively
> small. N exceeds the maximum dimension for a matrix, so I am looping to
> compute X'X and it takes ages.
>
> So, since this is the formula for the OLS beta=(X'X)^-1(X'Y), and
> "regress" is really fast I thought there must be an alternative way of
> computing X'X in a more efficient way.
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index