Thanks to Kit Baum updates for the -betafit-, -dirifit-, and
-propcnsreg- packages are now available from SSC. To install these
updates type -ssc install package_name, replace- (where package_name is
either betafit, dirifit, or propcnsreg), or -adoupdate, update-.
-betafit- allows one to estimate a regression on a depedent variable
which has a bounded range, for example a proportion which can by
definition only take values between 0 and 1. Mark Fossett pointed out a
bug in the way fweights are handled, this has been fixed.
-dirifit- is a multivariate generalization of -betafit-, instead of
modeling one proportion one can model multiple proportions which have
to add up to one, e.g. the proportion of a city budget spent on the
categories social services, security, roads, and other. Garry Anderson
pointed out that at the top of the output -dirifit- reports a Wald test
of the hypothesis that the coefficients of all variables in the first
equation are zero. This is the default for -ml-, but in case of
-dirifit- a more sensible hypothesis would be that the coefficients of
all variables in all equations are zero.
-propcnsreg- fits by maximum likelihood a linear regression with a
proportionality constraint. Consider a model where the explained
variable (y) depends on two explanatory variables (x1 and x2).
Furthermore suppose that the effects of x1 and x2 differ for different
values of a third variable (x3), i.e. the model contains interaction
effects between x1 and x3 and between x2 and x3. The proportionality
constraint imposes the constraint that the effects of x1 and x2 change
by the same proportion as x3 changes. So if the effect of x1 doubles
for a unit change in x3, than so does the effect of x2. Part of the
output is a test comparing the constrained and the unconstrained model.
In the last update I added a error by estimating the unconstrained
model without taking the -if- and -in- selections, and the weights into
account. This has been fixed.
-- 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/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
