James Rao <[email protected]>:
Note that second-order stochastic dominance (SOSD) is essentially the
same thing as generalized Lorenz dominance, so
findit glcurve
and see p.25 of http://stata-press.com/journals/stbcontents/stb48.pdf
for graphs, and see e.g.
http://www.mcgill.ca/files/economics/stochasticdominance.pdf
on inference. You might get better advice with more specifics on what
two distributions you want to analyze, and what your data looks like..
On Mon, Apr 20, 2009 at 9:00 AM, Nick Cox <[email protected]> wrote:
> In addition to Roger's excellent suggestions I'd underline the value of
> graphical comparison too, for example using -qqplot- and the ideas
> covered in
>
> SJ-7-2 gr0027 . . Stata tip 47: Quantile-quantile plots without
> programming
> . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . N.
> J. Cox
> Q2/07 SJ 7(2):275--279 (no
> commands)
> tip on producing various quantile-quantile (Q-Q) plots
>
> In essence, comparing the distribution functions graphically -- or
> equivalently the quantile functions -- should complement, and may tell
> you more than, any omnibus measure or test.
>
> Nick
> [email protected]
>
> Newson, Roger B
>
> I don't know what you mean by "2nd order stochastic dominance". However,
> stochastic dominance is usually measured using either Somers' D and
> Harrell's c. Confidence intervals for both of those can be calculated
> using the -somersd- package, downloadable from SSC using the -ssc-
> command. The -somersd- package is distributed with 3 .pdf manuals, which
> can be downloaded as ancillary files. The on-line help for -somersd-
> refers to some Stata Journal articles on the subject, which may be
> downloadable from the Stata Journal website at
> http://www.stata-journal.com/
>
> Rao, James
>
> I would like to do some stochastic dominance analysis to compae two
> distributions and I was just wondering if anyone knows how to conduct
> 1st and
> 2nd order stochastic dominance analysis in 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/
