Displaying 1 result from an estimated 1 matches for "bicvec".
Did you mean:
bichem
2010 Dec 26
1
Calculation of BIC done by leaps-package
...none
matches the calculation done by the summary-function. I already checked
for constants that could be the reason for the differences but i found
out, that the values vary apart of adding a constant term.
The source code of the leaps-package states the package calculates the
BIC this way:
bicvec<-c(bicvec,(n1+ll$intercept)*log(vr)+i*log(n1+ll$intercept))
with:
## number of observations - Intercept:
n1<-ll$nn-ll$intercept
## fraction of sum of squared residulas model i
## and sum of squared residuals null model, I
## just can't understand why the vector ll$ress
## is subscripted...