Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: RE: About stochastic dominance analysis??


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: About stochastic dominance analysis??
Date   Tue, 2 Dec 2008 14:35:30 -0000

There is perhaps some confusion here. Density functions must cross
somewhere, unless exceptionally they are identical. If that were not
true, so that one density function was everywhere higher than another,
then the densities could not both integrate over their respective
supports to 1, which is absurd. So crossing in itself is expected and
not in itself problematic. 

Jeetendra is perhaps confusing density functions with distribution
functions. 

On the latter, and in addition to other suggestions: see -distplot-
(SJ), -qplot- (SJ), -qqplot- (official Stata). Use -search- to find
locations. 

As Jeetendra has just two subsets, -qqplot- is a good direct approach to
the comparison of distribution functions. 

Nick 
n.j.cox@durham.ac.uk 

Jeetendra Aryal

I have two groups of farm households in my data, say Group A and Group
B. Group A used soil conservation measures while Group B did not. I have
data on yield (total value product per unit area) for each groups. I
used kdensity to examine whether yields on conserved plots dominates the
yield on non-conserved ones. However, the kdensity graphs are crossing
each other at some points and therefore, difficult to conclude.
Is there any way to do it?


*
*   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   |   What's new   |   Site index