Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Meta-analysis and corr2data


From   n p <nik_padazis@yahoo.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Meta-analysis and corr2data
Date   Tue, 9 Mar 2004 06:28:39 -0800 (PST)

Repeating the same procedure twice with different
seeds the results of the regression command were
identical at least for betas, SEs 95%CIs and t-values
(see output below) thus I suppose that one can use
corr2data to obtain such estimates.
Thanks to Richard Williams for the suggestions and the
useful link.

Nikos Pantazis

. corr2data x,n(10) mean(5) sds(0.5) seed(3)
. gen gr=1
. save cor1
. drop _all
. corr2data x,n(10) mean(10) sds(0.7) seed(3)
. gen gr=2
. save cor2
. drop _all
. corr2data x,n(10) mean(20) sds(1.2) seed(3)
. gen gr=3
. append using cor2
. append using cor1
. tab gr

. xi:reg x i.gr
i.gr              _Igr_1-3            (naturally
coded; _Igr_1 omitted)

      Source |       SS       df       MS             
Number of obs =      30
-------------+------------------------------          
F(  2,    27) =  802.75
       Model |  1166.66667     2  583.333335          
Prob > F      =  0.0000
    Residual |  19.6200016    27  .726666724          
R-squared     =  0.9835
-------------+------------------------------          
Adj R-squared =  0.9822
       Total |  1186.28667    29  40.9064369          
Root MSE      =  .85245

------------------------------------------------------------------------------
           x |      Coef.   Std. Err.      t    P>|t| 
   [95% Conf. Interval]
-------------+----------------------------------------------------------------
      _Igr_2 |          5   .3812261    13.12   0.000 
   4.217789    5.782211
      _Igr_3 |         15   .3812261    39.35   0.000 
   14.21779    15.78221
       _cons |          5   .2695676    18.55   0.000 
   4.446893    5.553107
------------------------------------------------------------------------------

. drop _all

. corr2data x,n(10) mean(5) sds(0.5) seed(15)
. gen gr=1
. save cor1,replace
. drop _all
. corr2data x,n(10) mean(10) sds(0.7) seed(15)
. gen gr=2
. save cor2,replace
. drop _all
. corr2data x,n(10) mean(20) sds(1.2) seed(15)
. gen gr=3
. append using cor2
. append using cor1
. xi:reg x i.gr
i.gr              _Igr_1-3            (naturally
coded; _Igr_1 omitted)

      Source |       SS       df       MS             
Number of obs =      30
-------------+------------------------------          
F(  2,    27) =  802.75
       Model |  1166.66667     2  583.333337          
Prob > F      =  0.0000
    Residual |  19.6200026    27  .726666764          
R-squared     =  0.9835
-------------+------------------------------          
Adj R-squared =  0.9822
       Total |  1186.28668    29  40.9064371          
Root MSE      =  .85245

------------------------------------------------------------------------------
           x |      Coef.   Std. Err.      t    P>|t| 
   [95% Conf. Interval]
-------------+----------------------------------------------------------------
      _Igr_2 |          5   .3812261    13.12   0.000 
   4.217789    5.782212
      _Igr_3 |         15   .3812261    39.35   0.000 
   14.21779    15.78221
       _cons |          5   .2695676    18.55   0.000 
   4.446893    5.553107
------------------------------------------------------------------------------



. 

--- Richard Williams <Richard.A.Williams.5@nd.edu>
wrote:
> At 01:48 AM 3/9/2004 -0800, n p wrote:
> >statistics etc.). The question is if it is right to
> >use the corr2data command to create three (or more)
> >independent samples for A, B and C arms with given
> >mean, SE and N to obtain via regression two betas
> and
> >two SEs (B vs. A and C vs. A). I know that I could
> use
> 
> Also, I discuss the use of -corr2data- for purposes
> like this on pp. 7-9 of 
> the following:
> 
>
http://www.nd.edu/~rwilliam/xsoc593/lectures/OLS-Stata.pdf
> 
> Next time I get around to revising that handout,
> I'll add the advice that, 
> if in doubt, generate alternative data sets with
> different seeds and see if 
> it makes a difference.  If it does make a
> difference, then what you are 
> trying to do is not ok.
> 
> 
> -------------------------------------------
> Richard Williams, Notre Dame Dept of Sociology
> OFFICE: (574)631-6668, (574)631-6463
> FAX:    (574)288-4373
> HOME:   (574)289-5227
> EMAIL:  Richard.A.Williams.5@ND.Edu
> WWW (personal):    http://www.nd.edu/~rwilliam
> WWW (department):    http://www.nd.edu/~soc
> 
> *
> *   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/


__________________________________
Do you Yahoo!?
Yahoo! Search - Find what you’re looking for faster
http://search.yahoo.com
*
*   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