Thanks to Kit Baum an update of the -propcnsreg- package is now
available on ssc.
-propcnsreg- estimates a linear regression with a
roportionality constraint using maximum likelihood.
The update now allows for an alternative identifying constraint.
One way to think about a proportionality constraint is that it
is estimating the effect of a latent variable, whereby the
latent variable is a weighted sum of the constrained variables.
Like any latent variable model one needs to impose constraints
to identify the scale and location of the latent variable. The
default is still to identify the latent variable by constraining
it to be zero when all its variables are at their minimum and to
one when all its variables are at their maximum. The alternative
is to constrain the a parameter or the sum of parameters to be
one. This way the latent variable is measured on the same scale
as the constrained (sum of) variable(s).
To update type -ssc install propcnsreg, replace-
-- Maarten
-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands
visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434
+31 20 5986715
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
*
* 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/
