Nick
The back transformation is not feasible because Z is a random variable and
E(exp(Z) is different from exp(E(z)).
The logit link is a veru good suggestion.
Thank you a lot
Eliana
>You can back transform algebraically using exp(Z) / (1 + exp(Z)), but
>another more general possibility is to use -glm, link(logit)-, which
>automatically yields predictions on the scale of Y. You need neither
>transformation nor back-transformation. There is still the question of
>which error family to use.
>
>See also
>How does one estimate a model when the dependent variable is a
>proportion?
>http://www.stata.com/support/faqs/stat/logit.html
>
>Nick
>[email protected]
>*
>* 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/
