I'm not sure what your ultimate purpose is, but another approach is to
extract the coefficients and build a simulator in Excel. You can make a
pretty slick tool by placing controls in a spreadsheet that allow you to
vary the values of each independent var and thereby set the corresponding
control's value, which are then used in formulas that automatically
calculate the predicted probabilities for you. I've used this to link the
probabilities to a pie chart that graphically shows the impact of
manipulating the independent vars. I find this infinitely more insightful
than staring at a table of coefficients and it's very easy to see the
influence of any number of combinations of independent variable values with
a few mouse clicks. It took some time to work out the kinks of this (the
biggest problem was making it an easy matter to adapt the spreadsheet to new
regression output), so I'd be happy to send you an example spreadsheet that
illustrates what I'm not describing too well.
-----Original Message-----
From: [email protected]
[mailto:[email protected]]On Behalf Of Daniel M�ller
Sent: Thursday, November 14, 2002 10:51 AM
To: Statalist
Subject: st: simulation of changes in RHS for -mlogit-
Hi,
probably a rather simple problem, but...
I would like to use the estimated coefficients from -mlogit- to calcuate new
outcome
probabilities for a dataset with some altered values of RHS variables.
I know how to access the coefficients stored in `e(b)' and that I can save
them in a
separate file. But how can I re-estimate the same model with known
coefficients without
explicitly writing the whole equation?
I would be very grateful for any hints,
best regards,
Daniel
*
* 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/
*
* 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/
