Hi everybody,I'm a beginner in R so this question it could may simple for u (I hope it).I need to study the interaction between 4 independent variables (x1, x2, x3, x4) and a dependent variable (y). So I've used a list of regression models (using the "lm" function) as:REGR.LIN<-lm(y~0+x1+x2+x3+x4)now I want to understand which could be the best model to use, using the "stepAIC" function :stepAIC(REGR.LIN, scope = list(lower=.~ 1, upper = .~(x1+x2+x3+x4)^4), , scale=4,k=2, step=1000)However, the point is that with this script I obtain:Start: AIC=3.240195e+19y ~ 0 + x1 + x2 + x3 + x4.....Step: AIC=2.771929e+19y ~ x1 + x2 + x3 + x4 + x1:x3 + x1:x4 + x2:x4 + x2:x3 + x1:x2 + x3:x4 + x1:x2:x3 + x1:x2:x4 + x2:x3:x4 + x1:x3:x4 + x1:x2:x3:x4 - 1There's a way to obtain all the other not linear combination (for example x1^4 : x2^4, x1^3 : x3^4 ...)????(I can have it when I write in the "lm" function all the variables, but I'm finding a way to obtain the same result in the stepAIC function, without changing anything in the lm function)Thank u!
-- View this message in context: http://r.789695.n4.nabble.com/stepAIC-with-lm-tp4685992.html Sent from the R help mailing list archive at Nabble.com. [[alternative HTML version deleted]] ______________________________________________ R-help@r-project.org 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.
