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: mi estimate and GLLAMM


From   "Pesola, Francesca" <[email protected]>
To   "[email protected]" <[email protected]>
Subject   RE: st: mi estimate and GLLAMM
Date   Fri, 4 Apr 2014 13:43:43 +0000

Thank you!

-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Stas Kolenikov
Sent: 04 April 2014 14:22
To: [email protected]
Subject: Re: st: mi estimate and GLLAMM

Re-impute that particular replicate (I heard some people call it implicate).

-- Stas Kolenikov, PhD, PStat (ASA, SSC)
-- Principal Survey Scientist, Abt SRBI
-- Opinions stated in this email are mine only, and do not reflect the position of my employer
-- http://stas.kolenikov.name



On Thu, Apr 3, 2014 at 10:17 AM, Richard Goldstein <[email protected]> wrote:
> it means exactly what it says: for the fourth set of imputed values 
> your model failed to converge; you might want to try running the model 
> on only the fourth set and trying to track down the problem that way
>
> Rich
>
> On 4/3/14, 11:03 AM, Pesola, Francesca wrote:
>> Hi Thanks for your reply. I am using Stata 11.
>>
>> I have added the cmdok option - below - :
>>
>> xi: mi estimate, cmdok: gllamm topic3g_w i.wave, i(ID) base(1) 
>> link(mlogit) family(binomial)
>>
>> However, Stata stops as I get an error message saying:
>>
>> convergence not achieved
>> model did not converge on m=4
>> r(498);
>> end of do-file
>> r(498);
>>
>> Any suggestion on why this might be and what to do?
>>
>> Cheers,
>> F
>>
>> -----Original Message-----
>> From: [email protected] 
>> [mailto:[email protected]] On Behalf Of Stas 
>> Kolenikov
>> Sent: 03 April 2014 15:14
>> To: [email protected]
>> Subject: Re: st: mi estimate and GLLAMM
>>
>> What version of Stata are you using? -mi estimate- has -cmdok- option that you need to specify to proceed with user-written commands. With some luck, -variance- option may give you the variances of the random effects, as by default -mi- would report the results in whatever metric -gllamm- uses to parameterize the random effects (Cholesky decomposition, if I recall correctly).
>>
>> -- Stas Kolenikov, PhD, PStat (ASA, SSC)
>> -- Principal Survey Scientist, Abt SRBI
>> -- Opinions stated in this email are mine only, and do not reflect 
>> the position of my employer
>> -- http://stas.kolenikov.name
>>
>>
>>
>> On Thu, Apr 3, 2014 at 5:30 AM, Pesola, Francesca <[email protected]> wrote:
>>> I have imputed my data using multiple imputation and I would now like to analyse the data using GLLAMM; however, Stata produces an error message which says GLLAMM is not supported by mi estimate.
>>>
>>> I know mi estimate supports the xt-family commands for clustered data but  there does not seem to be an xt command for multinomial logistic regression, which is what I need.
>>>
>>> Any suggestion would be very much appreciated!
>>>
>>> Thanks,
>>> Francesca
> *
> *   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/


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