st: Update to - domin - available on SSC

[Joseph Luchman <jluchman@gmail.com>]

Thanks to Kit Baum, a new version of - domin - has been uploaded to SSC.

- domin - version 2.0 is a substantial re-writing of the - domin -
program now utilizing Mata to increase speed and make the computation
of bootstrapped SE's feasible (within a reasonable time frame).

The major changes and updates are as follows:

a] computation of all n-tuples, all subset regression, and dominance
computations migrated to Mata (thanks to all individuals who pointed
out the errors the - tuples - command caused when interfacing with
domin)

b] incorporates stronger complete and conditional dominance criteria
(see helpfile for details)

c] ranking of predictors based on general dominance returned as a
matrix, e(ranking)

d] bug related to if and in qualifiers resolved - now if and in
qualifiers work as intended

e] dots representing each regression replaced with a progress bar for
predictors/sets>6 or >4 (for logits)

f] piechart dropped as option - I personally found this option to be
superfluous in retrospect

g] altered adjusted domin weight computation to result in
decomposition of adjusted r2's from full regression

h] incorporates "epsilon" or relative weights approach to general
dominance (for regress only)

i] McFadden's pseudo-R2 used for logit-based models (for consistency
with Azen & Traxel, 2009)

I welcome any and all comments from users and other interested parties
on the program, it's performance, as well as recommendations for
extensions/features.

To update type:

adoupdate domin, update

Note that - domin - requires Stata version 12.1 and Ben Jann's -
moremata - suite of Mata functions (available on SSC).

- joe

Joseph Nicholas Luchman, M.A.
----
Behavioral Statistics Lead | Fors Marsh Group
Email: jluchman@forsmarshgroup.com
forsmarshgroup.com
----
Doctoral Candidate
Industrial Organizational Psychology
George Mason University
https://www.researchgate.net/profile/Joseph_Luchman/
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/