Hello List Folks. I?m trying to do manova using a function which assembles
the response (LHS) and factors (RHS) from different sources. When I do so,
the model routines complain that I'm passing a list for the RHS when I
don't
think I am.
Here's a toy example:
r <- matrix(rnorm(30), ncol = 3) # toy data
td <- list(d = rnorm(10),
f1 = sample(c("A", "B"), 10, replace = TRUE),
f2 = sample(c("C", "D"), 10, replace = TRUE))
toyFn <-
function(my.list, my.data, fac = NULL) {
response <- my.data
fac <- data.frame(my.list[fac])
# str(fac) # a data frame!
man <- manova(response ~ fac)
print(summary(man))
invisible(man)
}
toyFn(td, r, fac = "f1")
toyFn(td, r, fac = c("f1", "f2"))
Bot attempts just above gives the following error:
Error in model.frame.default(formula = response ~ fac, drop.unused.levels TRUE)
:
invalid type (list) for variable 'fac'
But fac inside the function is a data frame. What am I missing here?
Also, I want the function to use this formula: response ~ f1:f2 but once I
get past my problem of data type, I don't think what I have so far will do
the interaction, in fact, I suspect it will use only f1 or perhaps f1 + f2,
just a hunch. Any suggestions to make this more elegant and robust? I've
been reading about model frames etc but don't quite see how to do a simple
example such as this.
Thanks, Bryan
*************
Bryan Hanson
Professor of Chemistry & Biochemistry
DePauw University, Greencastle IN USA
> sessionInfo()
R version 2.11.0 (2010-04-22)
x86_64-apple-darwin9.8.0
locale:
[1] en_US.UTF-8/en_US.UTF-8/C/C/en_US.UTF-8/en_US.UTF-8
attached base packages:
[1] datasets tools grid graphics grDevices utils stats
methods base
other attached packages:
[1] faraway_1.0.4 GGally_0.2 xtable_1.5-6 mvbutils_2.5.1
ggplot2_0.8.8
[6] proto_0.3-8 reshape_0.8.3 ChemoSpec_1.45 R.utils_1.4.0
R.oo_1.7.2
[11] R.methodsS3_1.2.0 rgl_0.91 lattice_0.18-5 mvoutlier_1.4
plyr_1.0.3
[16] RColorBrewer_1.0-2 chemometrics_0.8 som_0.3-5
robustbase_0.5-0-1 rpart_3.1-46
[21] pls_2.1-0 pcaPP_1.8-1 mvtnorm_0.9-9 nnet_7.3-1
mclust_3.4.4
[26] MASS_7.3-5 lars_0.9-7 e1071_1.5-23 class_7.3-2
