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

st: RE: RE: dominance testing

From   "Shehzad Ali" <>
To   <>
Subject   st: RE: RE: dominance testing
Date   Mon, 9 Jun 2008 13:57:34 +0100

Thanks, Nick. This is indeed written by World Bank folks and sadly the help
file is not available. I will try to contact the authors.



-----Original Message-----
[] On Behalf Of Nick Cox
Sent: 09 June 2008 13:46
Subject: st: RE: dominance testing

-dominance- is not found by -findit-. It appears to be a program
produced by World Bank employees. I can see no help file, always a
worrying sign. 

Such an error message might arise because the non-missing values for
your two variables are for quite different observations and the program
expects a different data structure. 
I'm guessing. 

I suggest you contact the authors for support. 


Shehzad Ali

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
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
dominance between two types of variables for the same individual?

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

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

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

*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index