No context and no rationale makes it
difficult to comment, except to say
that in models of type Y = Xb + e
normality of the individual X is
not a standard assumption. This is
not to say that logging might not
be a good idea on various grounds.
Nick
[email protected]
Ricardo Ovaldia
> Thanks to all that replied. Just to be clear the
> reviewer states "transformations
> (logarithmic or others as appropriate) are needed to
> normalize the distribution of skewed variables". This
> is clearly NOT the reason for wanting to transform
> independent variables in the Cox model.
*
* 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/
