David Harrison wrote:
Yes, why didn't I think of that form...?
Anyway, one other useful point to note. You can get a confidence
interval on the position of the maximum from the -nl- function (having
worked out how to express it as a function of the parameters) by using
-nlcom-:
nlcom ln(_b[/delta] / _b[/beta]) / (_b[/beta] - _b[/delta])
--------------------------------------------------------------------------------
The classic functional form jumps out from the scatterplot.
It still might be worthwhile to consider bootstrapping for the confidence
interval: -nl- and -nlcom- give asymptotic approximations.
Joseph Coveney
P.S. There was a stranded close-comment in one line
predict longevity_hat */
that should be deleted
predict longevity_hat
*
* 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/
