This is more of a statistics question than an R programming question. I suggest you look at Cross Validated for an answer. I found this in a quick search,
http://stats.stackexchange.com/questions/115571/back-transforming-regression-results-when-modeling-logy Jean On Thu, Dec 31, 2015 at 7:35 PM, Andras Farkas < fark...@optimum-dosing-strategies.org> wrote: > > > Dear All, > > wonder if you have a thought on the followimg: if I have a simple model > like model <- lm(log(y)~log(x)+log(z),data=data), where both, the dependent > and independent variables are log transformed, is it ok just to use ypred > <- predict(model,type=response) to get the predictions , then transform > ypred with exp(ypred) to y's original scale to compare observed or known > data (y) with model predicted (ypred) on the original scale? > > appreciate your thoughts... > > Andras > > ______________________________________________ > R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > [[alternative HTML version deleted]] ______________________________________________ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
