Hi,
I'm trying to select the best model for a particular problem. So far i
have managed to identify a set of variables that woudl explain my model
lm1 <- lm(Group ~ . , data=dataf))
> summary(lm1)
Df Sum Sq Mean Sq F value Pr(>F)
`A` 1 2.3963 2.3963 24.0390 7.328e-06 ***
`B` 1 0.7145 0.7145 7.1672 0.009525 **
`C` 1 0.6916 0.6916 6.9379 0.010680 *
`D` 1 5.7042 5.7042 57.2223 2.473e-10 ***
`E` 1 0.8928 0.8928 8.9563 0.003988 **
`F` 1 0.0036 0.0036 0.0359 0.850301
`G` 1 0.3817 0.3817 3.8295 0.054939 .
`H` 1 0.0581 0.0581 0.5833 0.447962
`I` 1 0.1371 0.1371 1.3756 0.245408
`J` 1 0.0504 0.0504 0.5052 0.479936
Residuals 61 6.0808 0.0997
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
I would like to run a stepwise analysis with the significant variables
from the previous lm. Can i input a filter to the stepAIC so that only
significant variables are used as the initial model to start step
analysis for ?
stepAIC(lm1,direction="both",) # How do i filter so that the initial
model start with the significant ones from lm testing.
thanks,
______________________________________________
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.
