Statalist The Stata Listserver


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

Re: RE: st: Regression: All possible subsets


From   Robert A Yaffee <bob.yaffee@nyu.edu>
To   statalist@hsphsun2.harvard.edu
Subject   Re: RE: st: Regression: All possible subsets
Date   Sun, 26 Nov 2006 00:18:16 -0500

Joe,
 Try ssc describe a
Then to install allpossible, type:
ssc install allpossible
- RY

Robert A. Yaffee, Ph.D.
Research Professor
Shirley M. Ehrenkranz
School of Social Work
New York University

home address:
Apt 19-W
2100 Linwood Ave.
Fort Lee, NJ
07024-3171
Phone: 201-242-3824
Fax: 201-242-3825
yaffee@nyu.edu

----- Original Message -----
From: Joseph Coveney <jcoveney@bigplanet.com>
Date: Saturday, November 25, 2006 9:48 pm
Subject: Re: RE: st: Regression: All possible subsets

> Raphael Fraser wrote:
> 
> tryem has no help file. I threw some variables at it and it froze.
> Nick?? Are you there?
> 
> --------------------------------------------------------------------
> ------------
> 
> Hmmm . . . -findit- doesn't bring up either -tryem- or -allpossible-.
> 
> -tryem- is an older module by Al Feiveson; -allpossible- is the one 
> by Nick
> Cox.  -tryem- does linear regression.  -allpossible- is more general.
> 
> In any event, there is a help file for -tryem- on my machine 
> (printed out
> below after the example of its use).  I'm not sure why it's not 
> there on SSC
> anymore.  The help file mentions that you'll need to have
> installed -mktime-.
> 
> Joseph Coveney
> 
> . clear
> 
> . set more off
> 
> . set obs 7
> obs was 0, now 7
> 
> . forvalues i = 1/7 {
>  2. generate float a`i' = 0.5 + 0.5 * (`i' == _n)
>  3. }
> 
> . mkmat a*, matrix(A)
> 
> . drawnorm y x1 x2 x3 x4 x5 x6, corr(A) n(25) seed(`=date("2006-11-
> 26","ymd")') clear
> (obs 25)
> 
> . forvalues i = 5(-1)1 {
>  2. tryem y x1 x2 x3 x4 x5 x6 `i'
>  3. }
> n = 6
> var =  y    x1    x2    x3    x4    x5    x6    5
> --------------------------------------------
> 
>  1  3  4  5  6
>    .300381  best of size 5
> subset is x1 x3 x4 x5 x6
> --------------------------------------------
> n = 6
> var =  y    x1    x2    x3    x4    x5    x6    4
> --------------------------------------------
> 
>  1  3  4  6
>    .299128  best of size 4
> subset is x1 x3 x4 x6
> --------------------------------------------
> n = 6
> var =  y    x1    x2    x3    x4    x5    x6    3
> --------------------------------------------
> 
>  1  4  6
>    .296403  best of size 3
> subset is x1 x4 x6
> --------------------------------------------
> n = 6
> var =  y    x1    x2    x3    x4    x5    x6    2
> --------------------------------------------
> 
>  1  6
>    .291414  best of size 2
> subset is x1 x6
> --------------------------------------------
> n = 6
> var =  y    x1    x2    x3    x4    x5    x6    1
> --------------------------------------------
> 
>  6
>    .201019  best of size 1
> subset is x6
> --------------------------------------------
> 
> . chelp tryem
> 
> --------------------------------------------------------------------
> ------------------------------------
> 
> Help file for the command tryem
> --------------------------------
> 
> Syntax
> ------
> 
> tryem var varlist k
> 
> 
> Options
> -------
> 
> Description
> -----------
> 'TRYEM': module to run all possible subset regressions
> tryem calculates all possible regressions of a given subset size.
> The maximum number of subsets, k, is one fewer than the total 
> number of
> independent variables.
> The module mktime is required.
> 
> tryem finds best regression subset of size k by brute force.
> It assumes independent variables are x1-xn and dependent variable 
> is y
> `1' = dep;   `2',..`nv+1' are indeps `nv+2'=k
> 
> 
> 
> Examples
> --------
> Example: tryem yyy xvar1 xvar2 xvar3 xvar4 2  finds best 2-subset 
> regressionof
> yyy on 4 indep variables xvar1,xvar2,xvar3,xvar4
> 
> 
> Author(s)
> ---------
> 
> Al Feiveson, Johnson Spaceflight Center
> Email: alan.h.feiveson1@jsc.nasa.gov
> 
> Also see
> --------
> 
> 
> --------------------------------------------------------------------
> ------------------------------------
> 
> .
> 
> *
> *   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