Re: st: New command -tfv- available from SSC

[daniel klein <klein.daniel.81@gmail.com>]

Austin,

-tfv- does neither of these things that we all would like to see.

I am terribly sorry for the misunderstanding caused by referencing
Richard's post. I did not mean to give the impression -tfv- is more
than it really is and can only repeat what I have stated already: the
command is just in the line of -xi-. No more, no less. And just as
-xi- allowed us to create interaction terms on the fly but did not
help us very much obtaining marginal effects of variables that are
part of those interaction terms, -tfv- does not help us obtain
marginal effects of transformed variables in their original metric.

Though most will already know, but something going in the suggested
direction is Patrick Royston's -mcp- (SJ)
http://www.stata-journal.com/article.html?article=gr0056.

Best
Daniel
-- 

And then computes marginal effects in terms of the untransformed
variable? With interactions?  I assume that is what people want, not a
command that takes the log on the fly.
*
*   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/