Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Obtaining 95%CI for marginal effect


From   Nur Hafidha Hikmayani <nhhikmayani@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Obtaining 95%CI for marginal effect
Date   Mon, 7 Feb 2011 11:48:19 +0700

To Shige & Richard, thanks for your suggestion for Stata 11 upgrading,
will consider it in the next few months.

To Steven,
First, thanks for pointing out the incorrect formula. It was intended
to estimate ME at medgrp=3, the command was supposed to be:
mfx, at(mean _Imedgrp_1=0 _Imedgrp_2=0 _Imedgrp_3=1).

Second, thanks again for the solution, it helps me a lot. Still have
few more questions anyway:
1. Is it at our discretion to adjust to rep78=3 (from your example)
considering perhaps that the 'risk' is higher at rep78=3, or was the
choice based on the largest proportion?
2. Why Stata gave error mesage when another numerical variable was added:
adjust weight=3138.575 turn=40.33816 _Irep78_2=0 _Irep78_3=1
_Irep78_4=0 _Irep78_5=0, by(foreign) ci se
Weighted means for weight & turn were previously estimated as well as
-svy:reg- for other dummy variables.

many thanks,
hafida-



On Sun, Feb 6, 2011 at 11:08 PM, Steven Samuels <sjsamuels@gmail.com> wrote:
>
> The method I showed will work for indicator variables too, but I, for one,
> don't understand what is represented when indicators for variables with >2
>  categories are set to their means.  (-svy: prop- won't help, as for k
> categories, it gives k proportions). Much better, I think, to set the other
> categorical variables  to  typical values. e.g., for the auto data set:
>
>  xi: svy: reg mpg weight i.foreign i.rep78
>  // adjust to rep78=3
>  adjust weight=3138.575 _Irep78_2=0 _Irep78_3=1 _Irep78_4=0 _Irep78_5=0,
> by(foreign)
>
> For non-linear models like -logistic-, setting variables to their means can
> produce unexpected results.  See:
>  http://www.stata.com/statalist/archive/2010-07/msg01596.html and Michael
> Norman Mitchell's follow-up.
>
>
> Note: Your original formulation in -mfx- looks incomplete.  You specified
> "at_Imedgrp_1=0 _Imedgrp_2=0" But medgrp had four levels,( 0,1,2,3), since
> -xi- produced three indicator variables. Your specification was equivalent
> to saying: at medgrp==0 or medgrp==3. Is this what you intended?
>
> Steve
>
>
> On Feb 6, 2011, at 9:59 AM, Nur Hafidha Hikmayani wrote:
>
> Thanks Steve.
> I'm afraid however that I'm not clear enough when some independent
> variables are categorical. You gave an example in which weight and
> turn are numerical variables - in my case, there is only 1 numerical
> IV. Suppose foreign is the main IV of interest and other covariates
> are mostly categorical, can we use -adjust- too (perhaps with
> -svy:prop- beforehand instead)?
>
> hafida-
>
> On Sun, Feb 6, 2011 at 9:12 PM, Steven Samuels <sjsamuels@gmail.com> wrote:
>>
>> Nur-
>>
>> I apologize. I checked and discovered -adjust- after -svy: reg- does not
>> compute predictions at the weighted means of the covariates, only at the
>> unweighted means. As a work-around, you could substitute the weighted
>> means
>> by hand.
>>
>> ****************************
>> sysuse auto, clear
>> drop if rep78==.
>> svyset rep78 [pw=head]
>> svy: mean weight turn   //get survey weighted means
>> xi: svy: reg mpg weight turn i.foreign
>> adjust weight= 3138.575 turn=40.33816, by(foreign) ci se
>> *****************************
>>
>> Steve
>> sjsamuels@gmail.com
>>
>>
>> Nur-
>>
>> Use -adjust- with the -ci- option.  The fitted value of y is not a
>> "marginal
>> effect";  for -regress- or (-svy: regress-) the default marginal effects
>> are
>> the regression coefficients.
>>
>> *********************
>> sysuse auto, clear
>> xi: reg weight price turn i.foreign
>> adjust price turn, by(foreign), se ci
>> ******************
>>
>> Steve
>> sjsamuels@gmail.com
>>
>> On Feb 6, 2011, at 7:02 AM, Nur Hafidha Hikmayani wrote:
>>
>> Dear all,
>> I've been running some regression models using -svy- and estimating
>> its marginal effect using -mfx- (I use Stata 10.1).
>> I wonder how can I get the 95% CI for the marginal effects (y)?
>>
>> The output for regression and its marginal effect are as follows:
>>
>> . xi: svy: reg GH i.medgrp exgrp chronic nummed gp
>>
>> ------------------------------------------------------------------------------
>>         |             Linearized
>>      GH |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
>>
>> -------------+----------------------------------------------------------------
>> _Imedgrp_1 |   4.429839    3.56263     1.24   0.214    -2.564628
>>  11.42431
>> _Imedgrp_2 |   8.333728   3.633545     2.29   0.022     1.200035
>>  15.46742
>> _Imedgrp_3 |   10.05818   3.773961     2.67   0.008     2.648813
>>  17.46755
>>   exgrp |  -8.916839    1.89046    -4.72   0.000    -12.62835   -5.205324
>>  chronic |  -10.31767   1.936802    -5.33   0.000    -14.12017   -6.515169
>>  nummed |  -1.063043   .3452676    -3.08   0.002    -1.740902   -.3851831
>>      gp |  -4.347845   1.773649    -2.45   0.014    -7.830027    -.865663
>>   _cons |   83.19335   3.520136    23.63   0.000     76.28231    90.10438
>>
>> ------------------------------------------------------------------------------
>>
>> . mfx, at(mean _Imedgrp_1=0 _Imedgrp_2=0)
>> Marginal effects after svy:regress
>>  y  = Fitted values (predict)
>>     =  53.085624
>>
>> ------------------------------------------------------------------------------
>> variable |      dy/dx    Std. Err.     z    P>|z|  [    95% C.I.   ]
>>  X
>>
>> ---------+--------------------------------------------------------------------
>> _Imedg~1*|   4.429839     3.56263    1.24   0.214  -2.55279  11.4125
>> 0
>> _Imedg~2*|   8.333728     3.63354    2.29   0.022   1.21211  15.4553
>> .379185
>> _Imedg~3*|   10.05818     3.77396    2.67   0.008   2.66136   17.455
>> 0
>> exgrp*|  -8.916839     1.89046   -4.72   0.000  -12.6221 -5.21161
>> .762312
>> chronic*|  -10.31767      1.9368   -5.33   0.000  -14.1137 -6.52161
>>  .83752
>> nummed |  -1.063043      .34527   -3.08   0.002  -1.73975 -.386331
>> 6.69376
>>  gp*|  -4.347845     1.77365   -2.45   0.014  -7.82413 -.871558   .472814
>>
>> ------------------------------------------------------------------------------
>> (*) dy/dx is for discrete change of dummy variable from 0 to 1
>>
>>
>> Any help is much appreciated,
>> Thanks,
>> hafida-
>>
>> *
>> *   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/
>>
>
> *
> *   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/
>

*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index