# Re: st: accessing delta-method-derived standard errors

 From "E. Paul Wileyto" To statalist@hsphsun2.harvard.edu Subject Re: st: accessing delta-method-derived standard errors Date Wed, 11 Jun 2008 12:24:14 -0400

```Check out the help on -nlcom-

You would get the exponentiated version of the SE by using :

-nlcom (exp([#2]_cons))

That will transform the constant asscoiated with the second ML equation.

To grab that value for other things, type -return list-

and you should see the matrices and scalars generated by nlcom

Paul

Feiveson, Alan H. (JSC-SK311) wrote:
```
Hi -
Some estimation commands (such as xtlogit below) give direct estimates
and standard errors for transformed parameters and then use the delta
method to show approximate standard errors for the corresponding orginal
parameters. My question is can one access these delta-method-derived
standard errors without doing the delta method by hand on actual
estimation standard errors?

For example, in the output below, I would like to retrieve the shown
standard error of sigma_u (1.092482) directly, rather than have to apply
the delta method to the standard error of log(sigma_u) (0.9960302).
Thanks

Al Feiveson

. xtlogit

Random-effects logistic regression Number of obs =
72
Group variable: id Number of groups =
12

Random effects u_i ~ Gaussian Obs per group: min =
6
avg =
6.0
max =
6

Wald chi2(1) =
8.44
Log likelihood = -29.679116 Prob > chi2 =
0.0037

------------------------------------------------------------------------
------
k8 | Coef. Std. Err. z P>|z| [95% Conf.
Interval]
-------------+----------------------------------------------------------
------
freq | 7.368662 2.536989 2.90 0.004 2.396254
12.34107
_cons | -3.744249 1.320748 -2.83 0.005 -6.332867
-1.155631
-------------+----------------------------------------------------------
------
/lnsig2u | 1.571154 .9960302 -.3810293
3.523338
-------------+----------------------------------------------------------
------
sigma_u | 2.193672 1.092482 .8265337
5.822145
rho | .5939471 .2402165 .171949
.9115323
------------------------------------------------------------------------
------
Likelihood-ratio test of rho=0: chibar2(01) = 7.55 Prob >= chibar2 =
0.003

. di _se[/lnsig2u]
.99603024

. di _se[/sigma_u]
r(111);

. di _se[/lnsig2u]sigma_u
invalid syntax
r(198);

*
* 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/

--
E. Paul Wileyto, Ph.D.
Assistant Professor of Biostatistics
Tobacco Use Research Center
School of Medicine, U. of Pennsylvania
3535 Market Street, Suite 4100

215-746-7147
Fax: 215-746-7140
epw@mail.med.upenn.edu
http://mail.med.upenn.edu/~epw/
*
* 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/