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

Re: st: RE: Returned results in gllamm


From   [email protected]
To   [email protected]
Subject   Re: st: RE: Returned results in gllamm
Date   Fri, 26 Aug 2005 09:16:21 +0100

Hi

e(chol) is the Cholesky decomposition of the variance matrix. 

See below: 

. matrix chol = (.97378367 , 0 \ -.86823121 , 3.586e-07)

. matrix chol2 = chol * chol'

. matrix list chol2

symmetric chol2[2,2]
            r1          r2
r1   .94825464
r2  -.84546937   .75382543

Tim 





"Mitchell, Michael" <[email protected]> 
Sent by: [email protected]
25/08/2005 21:04
Please respond to
[email protected]


To
<[email protected]>
cc
<[email protected]>
Subject
st: RE: Returned results in gllamm






Greetings
  I was searching for an answer to this very same question and found
Miguel's query, but no answer. I ran a random intercept, random slope
model with gllamm like this...

. eq cons : one
. eq x    : x
. generate one = 1
. gllamm y x , nrf(2) eqs(x cons) i(id) link(logit) family(bin) adapt
///
  nip(10)

and then got this output...

number of level 1 units = 2000
number of level 2 units = 500
 
Condition Number = 12.83709
 
gllamm model
 
log likelihood = -175.56749
 
------------------------------------------------------------------------
--
           y |      Coef.   Std. Err.      z    P>|z|   [95% Conf.
Interval]
-------------+----------------------------------------------------------
--
           x |     1.4471   .7456141     1.94   0.052  -.0142766
2.908477
       _cons |   3.038546   .6268254     4.85   0.000   1.809991
4.267101
------------------------------------------------------------------------
----
 
 
Variances and covariances of random effects
----------------------------------------------------------------------- 
***level 2 (id)
 
    var(1): .94825463 (.56235789)
    cov(2,1): -.84546937 (.95886005) cor(2,1): -1
    var(2): .75382543 (1.5231523)
----------------------------------------------------------------------

  I think I found the variance of the intercept this way... 

mat list e(chol)

e(chol)[2,2]
            c1          c2
r1   .97378367           0
r2  -.86823121   3.586e-07

. di .97378367 ^2 
.94825464

But this does not work for the variance of the slope...

. di 3.586e-07^2
1.286e-13

Any suggestions would be greatly appreciated!

Best regards,

Michael Mitchell
UCLA ATS Statistical Consulting Group
http://www.ats.ucla.edu/stat/ 


> -----Original Message-----
> From: [email protected] [mailto:owner-
> [email protected]] On Behalf Of miguel foguel
> Sent: Monday, July 25, 2005 10:32 AM
> To: [email protected]
> Subject: st: Returned results in gllamm
> 
> Dear all,
> 
> Could anyone tell me how I can retrieve the estimated variances and
> std-errors of the random "intercepts" from an estimation using
-gllamm-?
> They don't seem to be saved in the e(V) matrix.
> 
> Thank you in advance.
> 
> Miguel
> 
> 
> *
> *   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/


*
*   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 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index