Displaying 4 results from an estimated 4 matches for "rohwer".
2010 Aug 10
1
influence measures for multivariate linear models
...ted methods.
I'm interested in exploring the multivariate extension in R. I tried
the following, and was surprised to find that
R returned a result rather than an error -- presumably because mlm
objects are not trapped before they
get to lm.influence()
> # multivariate model
> data(Rohwer, package="heplots")
> rohwer.mod <- lm(cbind(SAT, PPVT, Raven) ~ n + s + ns + na + ss,
data=Rohwer)
> names(influence(rohwer.mod))
[1] "hat" "coefficients" "sigma" "wt.res"
> head(influence(rohwer.mod)$coefficie...
2012 Feb 09
1
passing an extra argument to an S3 generic
...lapse=',')
hat
}
where mlm.influence() does the calculations, but also allows the m=
argument to specify subset size.
Yet when I test this I can't seem to pass the m= argument directly, so
that it gets stuffed in to the infl=
call to mlm.influence.
# fit an mlm
library(heplots)
Rohwer2 <- subset(Rohwer, subset=group==2)
rownames(Rohwer2)<- 1:nrow(Rohwer2)
Rohwer.mod <- lm(cbind(SAT, PPVT, Raven) ~ n+s+ns+na+ss, data=Rohwer2)
> class(Rohwer.mod)
[1] "mlm" "lm"
## this doesn't work, as I would like it to, calling the hatvalues.mlm
method, b...
2010 Sep 14
0
influence measures for multivariate linear models
...S, and now conclude that there are bugs in lm.influence and the
coef(update()) methods I tried
before to get leave-one-out coefficients for multivariate response
models fit with lm().
By bugs, I mean that results returned are silently wrong, or at best,
misleading.
My test case: a subset of the Rohwer data anayzed by Barrett & Ling (&
others)
> library(heplots)
> data(Rohwer, package="heplots")
> Rohwer1 <- Rohwer[Rohwer$SES=='Hi',]
> rohwer.mod <- lm(cbind(SAT, PPVT, Raven) ~ n + s + ns + na + ss,
data=Rohwer1)
> (B <- coef(rohwer.mod))...
2013 Jan 29
3
how to suppress the intercept in an lm()-like formula method?
...t yet implemented") # lm.wfit(x, y, w, ...)
z$call <- cl
z$terms <- mt
z
}
TESTME <- FALSE
if (TESTME) {
# need to get latest version, 0.6-1 from R-Forge
install.packages("candisc", repo="http://R-Forge.R-project.org")
library(candisc)
data(Rohwer)
# this bombs: needs intercept removed
cc <- cancor.formula(cbind(SAT, PPVT, Raven) ~ n + s + ns + na + ss,
data=Rohwer)
## Error in chol.default(Rxx) :
## the leading minor of order 1 is not positive definite
#this works as is
cc <- cancor.formula(cbind(SAT, PPVT, Raven) ~ -1 + n + s +...