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: Neumark decomposition


From   Muhammad Anees <[email protected]>
To   [email protected]
Subject   Re: st: Neumark decomposition
Date   Tue, 17 Jul 2012 09:29:56 +0500

Hello,

I would recommend you read Jann (2008) paper very closely, reference
to which is given after the quote.

"An issue with the approach by Neumark (1988) or Oaxaca and Ransom
(1994) is that it may inappropriately transfer some of the unexplained
part of the differential into the explained component, although this
does not seem to have received much attention in the literature".

Quoted from: Jann, Ben (2008). The Blinder-Oaxaca decomposition for
linear regression models. The Stata Journal 8(4): 453-479.

I, thus, recommend you follow closely as to what is suitable for you.

Hope this helps.

Anees

On Tue, Jul 17, 2012 at 1:04 AM, Yingge Lin <[email protected]> wrote:
> Thanks so much Fernando. You are correct that I should have used
> -oaxaca- with omega for Neumark decompostion.
>
>
>
>
> On Mon, Jul 16, 2012 at 3:49 PM, Fernando Rios Avila <[email protected]> wrote:
>> Hi Yingge,
>> The syntax you are using for the -oaxaca- command is incorrect.
>> If you want to replicate the Neumark decomposition you need to use
>> oaxaca depvar [indepvars]  , by(groupvar) , omega
>> rather than "pooled"
>> According to the -oaxaca- help file, the pooled option includes the
>> excluded variable (in the syntax the ) in the pooled
>> regression. Whereas omega only includes the independent variables
>> declared.
>> Fernando
>>
>> On Mon, Jul 16, 2012 at 3:42 PM, Yingge Lin <[email protected]> wrote:
>>>
>>> Thank, Nick.
>>>
>>> The syntax I used is:
>>>
>>> .oaxaca depvar [indepvars]  , by(groupvar) , pooled
>>> and
>>> .decompose   depvar [indepvars], by(groupvar) detail estimates
>>>
>>> From my understanding, using pooled with _oaxaca will generate Neumark
>>> decompostion as it used the coefficient from the pooled data
>>> regression. And in the detail results that _decompose produced, there
>>> is one when D=* is also for Neumark decompostion. I run both program
>>> on the same data set but the results are not the same. I wonder
>>> whether I misunderstand the program or there is one program might
>>> contain error. i hope I could get some feedback from someone who also
>>> used those program and see whether same thing happen to them. Maybe
>>> someone on the list could help me and point out where I make mistake.
>>>
>>> Thanks again.
>>>
>>> Best,
>>> Yingge
>>>
>>>
>>>
>>> On Mon, Jul 16, 2012 at 3:18 PM, Nick Cox <[email protected]> wrote:
>>> > Yingge Lin:
>>> >
>>> > Rest assured that Maarten has a long track record of helping many,
>>> > many people on this list and also that he and Ben Jann are on entirely
>>> > good terms. Maarten did not explain his comment fully, but it is not
>>> > necessary to add to his remarks on the episode he mentioned. He was
>>> > merely trying to explain that Ben is no longer a member of this list,
>>> > so that it follows that you should ask Ben directly about his
>>> > programs. Even if users of these programs comment directly, that
>>> > remains good advice.
>>> >
>>> > The Statalist FAQ explains that you should assume goodwill even when
>>> > it is not evident. I would summarize Maarten's main advice as that it
>>> > is difficult to help you if you do not help yourself by asking precise
>>> > questions.
>>> > The purpose of such advice is not to shoot you down but to encourage
>>> > you to ask precise questions. Despite all that, it remains difficult
>>> > to comment on your experiences, including whether you have made
>>> > mistakes, as you give no details on exactly what command syntax you
>>> > used nor exactly what results you got.
>>> >
>>> >  I am commenting on your general aspersions; I don't use either of the
>>> > commands you mention.
>>> >
>>> > Nick
>>> >
>>> > On Mon, Jul 16, 2012 at 7:47 PM, Yingge Lin <[email protected]> wrote:
>>> >> Maarten
>>> >>
>>> >> Sorry if my question made you uncomfortable, but I think you should
>>> >> know that your answer has certainly made me very uncomfortable. I
>>> >> thought statalist is a place where people discuss the questions about
>>> >> STATA and learn from each other. All I wanted was some feedback from
>>> >> those who also have used -oaxaca- and -decompose-, and see whether I
>>> >> made mistake or one of the commands have some error. The -oaxaca- and
>>> >> -decompose- can be installed by ssc install.
>>> >>
>>> >> I do not know whether there is anything going on between Ben Jann and
>>> >> you, but it was never my intention to waste anybody’s time. But I do
>>> >> not appreciate that you tried to accuse me of looking for a shortcut
>>> >> only and wasting someone else time. It is untrue, and certainly if you
>>> >> felt so strongly that answering my questions was a waste of your time
>>> >> or were having a bad day, you really could have just chosen to ignore
>>> >> my questions.
>>> >>
>>> >> Yingge
>>> >>
>>> >> On Mon, Jul 16, 2012 at 12:13 PM, Maarten Buis <[email protected]> wrote:
>>> >>> On Mon, Jul 16, 2012 at 5:48 PM, Yingge Lin wrote:
>>> >>>> I am using STATA10. Ben Jann also have -oaxaca- for earlier version of
>>> >>>> stata (8 and 9). Hope these clarify a little bit of my questions.
>>> >>>
>>> >>> That was not what I was asking. I asked where you got the user written
>>> >>> software from (-oaxaca- and -decompose-), not which version of Stata
>>> >>> you were using. However, you should also have told use which version
>>> >>> of Stata you were using. Since you did not say so in your original
>>> >>> question I assumed, per the Statalist FAQ, you used the most recent
>>> >>> (12).
>>> >>>
>>> >>> I realize that often we want answers quickly, and thus take
>>> >>> short-cuts, like not reading the FAQ. But this illustrates that short
>>> >>> cuts often lead to more work. If this only meant that you wasted your
>>> >>> own time, than that is not so bad: hopefully you have learned
>>> >>> something in the process. However, you are also wasting our time.
>>> >>> Since you rely on our good will to get answers, wasting our time is
>>> >>> not a good idea. So I strongly suggest that you take your time to read
>>> >>> the FAQ carefully before posting anything on the list.
>>> >>>
>>> >>> Anyhow, my answer remains the same: Ben Jann is no longer active on
>>> >>> the list (a sad story, involving one of the most dysfunctional periods
>>> >>> on statalist), so you need to contact him directly.
>>> >>>
>>> >
>>> > *
>>> > *   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/



-- 

Best
---------------------------
Muhammad Anees
Assistant Professor/Programme Coordinator
COMSATS Institute of Information Technology
Attock 43600, Pakistan
http://www.aneconomist.com

*
*   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