>>Good morning, >> >>I am a PhD student in Barcelona working with R. My question is about the >>summary of linear regressions. >>The output of that summary gives some statistical parameters of the >>regression. One of them is the R-squared. In the help menu i have read >>that the manner to calculate the R-squared is >> >>R^2 = 1 - Sum(R[i]^2) / Sum((y[i]- y*)^2), and when the linear >>regression is forced to start at the point (0,0) y* is 0. >> >>My question is if the R-squared obtained when the linear regression >>crosses the origin has the same meaning as when it does with an >>intercept. I'm faced to a theoretical model which brings some values for >>a variable that compared to the observed ones gives some R*squared of >>0.32 when the regression has an intercept and 0.7 when it is forced to >>cross the origin. >>To sum up, it may be stated that the theoretical methodolgy followed >>explains the 70% of of the observed variance? >> >>Thank you very much. >> >>Gorka Merino.
Gorka Merino Institut de Ciències del Mar, CMIMA-CSIC Psg. Marítim de la Barceloneta 37-49 08003-BARCELONA (Spain) Tel.: (34) 932 30 95 48 e-mail: [EMAIL PROTECTED] CMIMA: Tel.: (34) 932 30 95 00 Fax: (34) 932 30 95 55 ______________________________________________ R-help@stat.math.ethz.ch mailing list 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.