st: RE: Regressions with combinatorial of variables

["Keith Dear" <Keith.Dear@anu.edu.au>]

José,
I think this does what you want:


* set up some data with weak relationships
glo N 131			// number of x-variables
set obs 256
forval i=1/$N {
	gen var`i'=runiform()
}
egen Y = rsum(var*)		// outcome variable
replace Y = 0.2*Y + rnormal(0,1)

* now the work. $X holds a growing list of significant vars
glo X
forval i=1/$N {
	di "`i': $X"
	qui reg Y $X var`i'
	qui test var`i'
	if r(p)<.05 glo X $X var`i'
}




-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of José Luis Chávez Calva
Sent: Wednesday, 2 February 2011 6:03 AM
To: statalist@hsphsun2.harvard.edu
Subject: st: Regressions with combinatorial of variables

Hi,

I have to run different regressions with combinations of variables (I
have 131 vars), the problem is the following

I have to run:
    reg var1
case 1) var1 is significant, then
    reg var1 var2
case 2) car1 is not significant, then
    reg var2
In both cases, if var2 is significant, then
case 1) reg var1 var2 var3
case 2) reg var2 var3
In both cases, if var2 is not significant, then
case 1) reg var1 var3
case 2) reg var3
....
And so on, in fact it is a kind of tree, I have tried different
especifications (forvalues, locals, if,etc.) but I didn´t have
results, I'm wondering if some of you have worked with something like
this.


Thanks


best regards



José Luis

p.s. It is for obtain a smoother of the serie.

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/