Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


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

Re: st: sem - is it possible to drop error terms?


From   Jacob Felson <felsonj@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: sem - is it possible to drop error terms?
Date   Tue, 12 Mar 2013 16:39:35 -0400

My syntax is below.  All covariances between latent variables should
be set to 0 *except* the covariance between A1 and A2, which is set to
1 for identical (MZ) twins and 0.5 for fraternal (DZ) twins, and the
covariance between C1 and C2, which is set to 1.



#delimit ;
sem (depression123_1 <- A1@a C1@c E1@e) (depression123_2 <- A2@a C2@c
E2@e) if inlist(sibtype,1,2),
var(e.depression123_1@0 e.depression123_2@0)
cov(e.depression123_1*e.depression123_2@0) cov(1: A1*A2@1) cov(2:
A1*A2@.5) cov(C1*C2@1) cov(A1*C1@0) cov(A1*C2@0) cov(A2*C1@0)
cov(A2*C2@0) cov(E1*A1@0) cov(E1*A2@0) cov(E1*A3@0) cov(E2*A1@0)
cov(E2*A2@0) cov(E2*C1@0) cov(E2*C2@0)
group(sibtype) nomeans noconstants ginvariant(mcoef mcons)
;


A path diagram of this type of model is shown in the image at the
link: http://www.biomedcentral.com/1471-2202/5/49/figure/F1?highres=y.





On Tue, Mar 12, 2013 at 4:05 PM, John Antonakis <John.Antonakis@unil.ch> wrote:
> If you show your syntax maybe someone could help you.
>
> Best,
> J.
>
> __________________________________________
>
> John Antonakis
> Professor of Organizational Behavior
> Director, Ph.D. Program in Management
>
> Faculty of Business and Economics
> University of Lausanne
> Internef #618
> CH-1015 Lausanne-Dorigny
> Switzerland
> Tel ++41 (0)21 692-3438
> Fax ++41 (0)21 692-3305
> http://www.hec.unil.ch/people/jantonakis
>
> Associate Editor
> The Leadership Quarterly
> __________________________________________
>
> On 12.03.2013 20:17, Jacob Felson wrote:
>>
>> Stas,
>>
>> Thanks!  Its strange -- when I try to run a basic behavior genetic
>> model (ACE) using Stata's sem command, I immediately receive an error
>> saying there are too many latent variables and the model is not
>> identified.  But this model is definitely identified -- at least
>> theoretically.  Model is here:
>> http://www.biomedcentral.com/1471-2202/5/49/figure/F1?highres=y
>>
>> Sorry to bother you.  Any help you could provide is much appreciated.
>>
>> Sincerely,
>>
>> Jacob Felson
>>
>>
>>
>> On Tue, Mar 12, 2013 at 2:59 PM, Stas Kolenikov <skolenik@gmail.com>
>> wrote:
>>>
>>> Jacob,
>>>
>>> you can constrain the variance of these error terms to zero with
>>> -variance()- option of -sem-.
>>>
>>> -- Stas Kolenikov, PhD, PStat (SSC)
>>> -- Senior Survey Statistician, Abt SRBI
>>> -- Opinions stated in this email are mine only, and do not reflect the
>>> position of my employer
>>> -- http://stas.kolenikov.name
>>>
>>>
>>>
>>> On Tue, Mar 12, 2013 at 1:53 PM, John Antonakis <John.Antonakis@unil.ch>
>>> wrote:
>>>>
>>>> Hi Jacob:
>>>>
>>>> You need to be much more specific about what specific model you wish to
>>>> test
>>>> (write-out equations or syntax). Most probably don't know what ACE is:
>>>>
>>>> http://www.stata.com/support/faqs/resources/statalist-faq/
>>>>
>>>> Best,
>>>> J.
>>>>
>>>> __________________________________________
>>>>
>>>> John Antonakis
>>>> Professor of Organizational Behavior
>>>> Director, Ph.D. Program in Management
>>>>
>>>> Faculty of Business and Economics
>>>> University of Lausanne
>>>> Internef #618
>>>> CH-1015 Lausanne-Dorigny
>>>> Switzerland
>>>> Tel ++41 (0)21 692-3438
>>>> Fax ++41 (0)21 692-3305
>>>> http://www.hec.unil.ch/people/jantonakis
>>>>
>>>> Associate Editor
>>>> The Leadership Quarterly
>>>> __________________________________________
>>>>
>>>>
>>>> On 12.03.2013 19:30, Jacob Felson wrote:
>>>>>
>>>>> I'm wondering if its possible to drop the error terms that are
>>>>> included by default in sem models.  I've searched in the manual and
>>>>> online but I can't find any indication that this is possible.  I
>>>>> imagine its not, but I wanted to check with the experts to make sure.
>>>>>
>>>>> (I'm estimating a behavior genetic model with data on MZ and DZ twins
>>>>> -- the so-called ACE model.)
>>>>>
>>>>>
>>>>> Thank you,
>>>>>
>>>>>
>>>>> Jacob Felson
>>>>> *
>>>>> *   For searches and help try:
>>>>> *   http://www.stata.com/help.cgi?search
>>>>> *   http://www.stata.com/support/faqs/resources/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/faqs/resources/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/faqs/resources/statalist-faq/
>>> *   http://www.ats.ucla.edu/stat/stata/
>>
>>
>>
>> --
>> Jacob Felson
>> *
>> *   For searches and help try:
>> *   http://www.stata.com/help.cgi?search
>> *   http://www.stata.com/support/faqs/resources/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/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/



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


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index