Kit Baum replied:
> Quite so, but automatically transforming variables by ln(x/(1-x))
> assumes that they are in a proper interval (such as inside (0,1)) in
> the first place. I'm not sure how Clive's approach, using the
> unemployment rate (a value always in excess of 2) works very well
> with a logit transformation. Being able to automatically generate a
> large number of variables with all missing values is probably of
> limited utility.
I was desperately trying to search for a toy dataset that any Statalister
could access which contained %-level variables. Not many do, I notice, but
productivity.dta was one of them. Unfortunately, the normalized -unemp-
variable, although S-shaped, generates a rather flat distribution:
. g nunemp=unemp/100
. twoway (line nunemp id), ylabel(0(.1)1)
Still, it doesn't cross its bounds. Had I noticed this before posting, I'd
have searched for a better-behaved example from Stata's data library.
CLIVE NICHOLAS |t: 0(044)7903 397793
Politics |e: [email protected]
Newcastle University |http://www.ncl.ac.uk/geps
Whereever you go and whatever you do, just remember this. No matter how
many like you, admire you, love you or adore you, the number of people
turning up to your funeral will be largely determined by local weather
conditions.
*
* 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/
