Statalist


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

st: RE: RE: dominance testing


From   "Verkuilen, Jay" <[email protected]>
To   <[email protected]>
Subject   st: RE: RE: dominance testing
Date   Mon, 9 Jun 2008 09:54:36 -0400

Nonparametric ROC analysis is also an ordinal dominance procedure, and
it's in official Stata (-roctab-). Donald Bamber's Journal of Math Psych
paper (cited in the -roctab- help) is posed as a question of stochastic
dominance, not as ROC analysis at all. I have no idea if it does what is
needed here but it's worth a look. 

Jay

-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of JP Azevedo
Sent: Monday, June 09, 2008 9:49 AM
To: [email protected]
Subject: st: RE: dominance testing


Hello Shehzad,

Please notice that that the user writtem command alorenz also test the
stocastist dominance (first order, second order and Lorenz dominance).

ssc alorenz, install


    compare Analysis the first order stochastic dominance (Saposnik,
1981, 1983), second order
    stochastic dominance (Shorrocks, 1983), and the Lorenz dominance of
the distribution (Atkinson,
    1970).

Best, JP



-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Shehzad Ali
Sent: Monday, June 09, 2008 2:03 AM
To: [email protected]
Subject: st: dominance testing

Hi,

I am trying to test dominance of concentration curves using - dominance
- command. Say there are 1,000 individuals in the sample and 300 of them
have insurance and 700 don't. Concentration curves of their health care
expenditures are plotted using - glcurve - command. Is it legitimate to
test dominance of the two curves against each other or do we only test
for dominance between two types of variables for the same individual?

When I run the dominance command I get an error saying 'no
observations'. 
Here is what I am doing:

dominance y1 y2 [aw=wt], sortvar(x) rule(both)

where,
y1= health expenditure of insured
y2=health expenditure of uninsured
x=income variable

Any help would be appreciated.

Cheers,

Shehzad
*
*   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/

*
*   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