Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: How I can run a tree regression model in Stata


From   "Quang Nguyen" <[email protected]>
To   [email protected]
Subject   Re: st: How I can run a tree regression model in Stata
Date   Mon, 11 Dec 2006 18:57:23 -1000

Dear Richard et al,

Thanks so much again for your comments. To clarify the matter a bit,
the data is collected from an experimental economics. Each participant
is randomly assigned as player 1, 2, or 3. Then s/he is given a piece
of paper and ased to make the decision. For player 1 It is something
like this:

If Player 2 is in White Group and Player 3 is in Black Group,
I will vote for Plan 12 / Plan 13 / Plan 23.    (Please circle one)

If Player 2 is in White Group and Player 3 is in Pink Group,
I will vote for Plan 12 / Plan 13 / Plan 23.     (Please Circle one)

and so on.

Your suggestion on the interaction effect is a good one and we can do that.

Our data have information on the participant chracteristics such as
age, education, income etc. We would like to see how these
characteritics may have impact on the decision they make under these
above scenaria. Could you give me a brief idea of how I can run with
MNL model? Is it just like a regular MNL?

Thanks so much and Have A Great Day!






On 12/11/06, Richard Williams <[email protected]> wrote:
At 11:14 PM 12/11/2006, Quang Nguyen wrote:
>Dear Richard el al,
>
>Thanks so much for your reply. You raised an excellent question.
>Actually, the colors represents the characteristics of the players by
>ethnicity and income level. For instance, "white" represents for
>Rich_Vietnamese, "black" represents for Poor_Vietnamese and so on. The
>main research question is to see how individual's perception on income
>an dethnicity might have effect on the "cooperation/coordinating"
>behavior. For instance, the rich might be more willing to cooperate
>with the poor for upward social comparision reason.

To further clarify - you originally said players are "required" to
vote based on the characteristics of others.  But, the above makes it
sound like they are not required to do so.  Rather, you want to test
whether the ethnic and economic characteristics of others will affect
the vote?  Or, perhaps you could say they are required to vote based
on characteristics, but it is up to them to decide how to do so.

The color coding seems to confound the effects of race and
income.  For each of the other 2 players, I'd be inclined to code
their race and income separately.  You could then compute
interactions if you thought it might be appropriate or necessary.

I'm still not clear on how to set everything up, but I definitely
don't see how this would be a regression problem.  What would the
continuous dependent variable be?  Every possible dv presented so far
has 3 categories.

-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX:    (574)288-4373
HOME:   (574)289-5227
EMAIL:  [email protected]
WWW (personal):    http://www.nd.edu/~rwilliam
WWW (department):    http://www.nd.edu/~soc

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index