Stephen - It sounds as if you really have a mixed discrete-continuous
distribution with an atom at zero. I don't see why the distribution would
bimodal, however. Nevertheless, you can't simply take logs if zeros are in
your data. Perhaps you need something like a -heckman- regression model
where a "zero" is considered a non-response. Then you could apply this model
to logs of the non-zero values.
Al Feiveson
-----Original Message-----
From: [email protected]
[mailto:[email protected]]On Behalf Of Stephen Soldz
Sent: Wednesday, June 02, 2004 1:53 PM
To: [email protected]
Subject: st: Log transform of skewed data
I have data on the "cost" (actually tranformed hours) of various types of
caretaking for Alzheimers patients. I'm interested in a regression model to
test treatment effects in a multisite study. As is usual for cost data, it
is positively skewed. So, I contemplated a log transform, either through a
direct transformation of the response, or through a log link in a glm, gee,
or something similar. I actually am using "xt" commands to allow for
nonindependence among caretakers treated at the same site.
the problem is that the mode cost is $0, so that the distribution is
bimodal. This, of course, remains true if I do a lof transform. Any ideas on
how to analyze such data would be apreciated.
Stephen Soldz
Director, Center for Research, Evaluation, and Program Development
Boston Graduate School of Psychoanalysis
1581 Beacon St.
Brookline, MA 02446
[email protected]
http://soldzresearch.com/stephensoldz
*
* 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/
