The R-squared that you get after transforming Y is a measure of the how well the model predicts the transformed Y. ie, how much of the % variability in log(y) is explained by the model.
But, the R-squared of the two models:
y= b0 + b1*x and
log(y) = b0 + b1*x , cannot be compared with each other since the independent variables are not the same.
Hope this helps.
Leny
*
* 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/
