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: Questions about selmlog and svyselmlog


From   "Yu(Sophia) Zhang" <sophia.zhangyu@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Questions about selmlog and svyselmlog
Date   Thu, 21 Oct 2010 14:52:34 -0400

Dear Joao,

   Thanks for your prompt reply.

   The correct data structure will make me loose most of the values of
depvar. What's your comments on this? Do I need to run 3 seperate
selmlog regression?

   Thanks

   Yu

On Thu, Oct 21, 2010 at 2:45 PM, Joao Ricardo F. Lima
<jricardofl@gmail.com> wrote:
> Yu,
>
> your data structure is wrong. This is the correct way:
>
>>   depvar    varlist, select ( depvar_m  =   varlist_m)....
>>
>>    500          x                        1                      z
>>    456          x                        1                      z
>>    529          x                        1                      z
>>    601          x                        1                      z
>>    .         x                        2                      z
>>    .          x                        2                      z
>>    .          x                        2                      z
>>    .          x                        3                      z
>>    .          x                        3                      z
>>    .          x                        3                      z
>>    .         x                        3                      z
>>    .         x                        3                     z
>>
>
> Joao Lima
>
>
> 2010/10/21 Yu(Sophia) Zhang <sophia.zhangyu@gmail.com>:
>> Dear Joao,
>>
>>   Thanks for your patient explanation. I thought I understood what
>> you mean. But when I followed the rule and made sure that the
>> dependent variable has non-missing values for only one category, the
>> same error still showed up.
>>
>>    What I understand the correct data structure about missing values
>> of depvar is like this:
>>
>>   depvar    varlist, select ( depvar_m  =   varlist_m)....
>>
>>    500          x                        1                      z
>>    456          x                        1                      z
>>    529          x                        1                      z
>>    601          x                        1                      z
>>    miss         x                        2                      z
>>    547          x                        2                      z
>>    408          x                        2                      z
>>    620          x                        3                      z
>>    599          x                        3                      z
>>    512          x                        3                      z
>>    miss         x                        3                      z
>>    miss         x                        3                     z
>>
>>    So the depvar has non-missing values for category 1 only. It has
>> missing values for all the other categories (2 and 3).
>>
>>    Would you please kindly correct me if I am wrong?
>>
>>    I appologize to ask another question. If the command works, how
>> could save the variable lambda (Heckman correction term)?
>>
>>    I appreciate your help a lot!
>>
>>    Yu
>>
>>
>>
>>
>> On Wed, Oct 20, 2010 at 11:08 AM, Joao Ricardo F. Lima
>> <jricardofl@gmail.com> wrote:
>>> Dear Yu,
>>>
>>> the sintax of selmlog is:
>>>
>>> selmlog depvar varlist, select(depvar_m=varlist_m) [lee dmf(#) dhl(# [all]) etc
>>>
>>> depvar_m is your categorical variable, with more than 2 categories, oks?
>>>
>>> and
>>>
>>> depvar must have non-missing values for only one of theses categories.
>>>
>>> HTH,
>>>
>>> JL
>>>
>>>
>>>
>>> 2010/10/20 Yu(Sophia) Zhang <sophia.zhangyu@gmail.com>:
>>>> Dear Joao,
>>>>
>>>>  Thanks so much for your reply. The second question has been solved
>>>> following your suggestion.
>>>>
>>>>   For the first one, do you mean that across all categories ( which
>>>> are schools in my selection equation), only one category can have
>>>> non-missing values of dep var, and the other categories must have some
>>>> missing values of depvar? I deleted some values of depvar to meet this
>>>> requirement, but I still got the same error. Would you mind explaining
>>>> it in more detail?
>>>>
>>>>  Thank you so much!
>>>>
>>>>  Yu
>>>>
>>>> On Tue, Oct 19, 2010 at 7:03 PM, Joao Ricardo F. Lima
>>>> <jricardofl@gmail.com> wrote:
>>>>> Yu,
>>>>>
>>>>> About the first question:
>>>>>
>>>>> The vardep  of the income equation must have non missing values ONLY
>>>>> for one category.
>>>>>
>>>>> About the second:
>>>>>
>>>>> the svyselmlog was written to Stata 8. Then, try this:
>>>>>
>>>>> version 8: svyset [pweight = varw], psu(varpsu) strata(varstrat)
>>>>> svyselmlog vardep varindep , sel() meth() boot()....
>>>>>
>>>>> Hope this Helps,
>>>>>
>>>>> Joao Lima
>>>>>
>>>>>
>>>>> 2010/10/19 Yu(Sophia) Zhang <sophia.zhangyu@gmail.com>:
>>>>>> Dear statisticians,
>>>>>>
>>>>>>     I am now working on my dissertation and am stuck by a STATA command.
>>>>>> Basically, I want to run Heckman correction based on the multinomial logit
>>>>>> model. I am going to use the command selmlog. However, a strange error shows
>>>>>> up like this:
>>>>>>
>>>>>> Non-missing values of dependent variable for more than one outcome
>>>>>> variable _m* not found
>>>>>>
>>>>>>    I guess it might be associated with this statement in the help file:
>>>>>>
>>>>>> 1) In the output, selmlog adds to varlist a series of variables labelled
>>>>>> _m[depvar_m value], except for the dhl option
>>>>>>
>>>>>> 2) The outcome variable depvar is observed for only one value of depvar_m.
>>>>>> It is important that depvar should have missing values for any other value
>>>>>> of depvar_m.
>>>>>>
>>>>>>    But still cannot figure out what's wrong.
>>>>>>
>>>>>>    In addition, when I want to control for sampling weights by running the
>>>>>> command svyselmlog, I get another error:
>>>>>>
>>>>>> data not set up for the old survey estimation commands; use svyset under
>>>>>> version control or use the new svy prefix.
>>>>>>
>>>>>>   But there is no "svy: selmlog" command in STATA 11.....
>>>>>>
>>>>>>     I wonder if someone could kindly give me some hint for these two
>>>>>> problems. I have asked several people but no one were familiar with this
>>>>>> problem.
>>>>>>
>>>>>>     I appreciate your help very much.
>>>>>>
>>>>>>     Sincerely,
>>>>>>
>>>>>>     Yu
>>>>>>
>>>>>> *
>>>>>> *   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/
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> ---------------------------------------------------------
>>>>> João Ricardo F. de Lima
>>>>> Pesquisador A - Economia Aplicada
>>>>> Embrapa Semiárido
>>>>> e-mail: joao.ricardo@cpatsa.embrapa.br
>>>>> fone: +55 (87) 3862-1711 Ramal: 167
>>>>> home-page: www.cpatsa.embrapa.br
>>>>> ----------------------------------------------------------
>>>>>
>>>>> *
>>>>> *   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/
>>>>>
>>>>
>>>>
>>>>
>>>> --
>>>> Yu Zhang
>>>> PhD Candidate
>>>> Economics and Education
>>>> Teachers College, Columbia University
>>>>
>>>> *
>>>> *   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/
>>>>
>>>
>>>
>>>
>>> --
>>> ---------------------------------------------------------
>>> João Ricardo F. de Lima
>>> Pesquisador A - Economia Aplicada
>>> Embrapa Semiárido
>>> e-mail: joao.ricardo@cpatsa.embrapa.br
>>> fone: +55 (87) 3862-1711 Ramal: 167
>>> home-page: www.cpatsa.embrapa.br
>>> ----------------------------------------------------------
>>>
>>> *
>>> *   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/
>>
>
>
>
> --
> ---------------------------------------------------------
> João Ricardo F. de Lima
> Pesquisador A - Economia Aplicada
> Embrapa Semiárido
> e-mail: joao.ricardo@cpatsa.embrapa.br
> fone: +55 (87) 3862-1711 Ramal: 167
> home-page: www.cpatsa.embrapa.br
> ----------------------------------------------------------
>
> *
> *   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/
>



-- 
Yu Zhang
PhD Candidate
Economics and Education
Teachers College, Columbia University

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