[R] Strange R squared, possible error

derek jan.kacaba at gmail.com
Fri Mar 18 00:20:31 CET 2011


Thank you for our reply. It's a pity, that 2 variables defined by different
formula have same name. If the variables had been named differently, I
wouldn't have problem at all and it looks like it's done on purpose.
Because I test a quality of data (performance of collecting data) not a
model which I know a priori, R^2 gives me simple information of that quality
even for nonlinear model.

Still when testing different models R^2 shows nothing important. I didn't
stated that negative R^2 is meaningless. It has meaning that the model or
the data are wrong.

--
View this message in context: http://r.789695.n4.nabble.com/Strange-R-squared-possible-error-tp3382818p3385837.html
Sent from the R help mailing list archive at Nabble.com.



More information about the R-help mailing list