Hello All..
Please consider the following:
y <- rnorm(20, mean = 10)
f1 <- as.factor(rep(c("A", "B", "B",
"A"), 5))
f2 <- as.factor(rep(c("C", "D"), 10))
testdata <- data.frame(y, f1, f2)
testFunc <- function(formula, data, ...) {
# mf <- model.frame(formula, data)
kw.res <- kruskal.test(formula, data)
}
res <- testFunc(y ~ f1 * f2, testdata)
Which works perfectly well. Now, I would like to do some other tests which
allow only one factor, for instance wilcox.test. So, I would like to modify
my formula so that it reads y ~ f1 or y ~ f2, then conduct more tests. What
is the smart way to subset or modify an existing symbolic formula? I have
been staring at the attributes of res and mf (commented out currently) but
these are nested in a way I find difficult to extract. ?terms,
?terms.formula, ?formula etc all discuss these attributes but as I said, it
seems a bit impenetrable. RSiteSearch("symbolic formula") returns too
many
answers. No doubt I am missing the obvious, as this is my first foray into
using formulas in my own functions.
TIA, Bryan
*************
Bryan Hanson
Acting Chair
Professor of Chemistry & Biochemistry
DePauw University, Greencastle IN USA
> sessionInfo()
R version 2.10.0 RC (2009-10-19 r50172)
i386-apple-darwin8.11.1
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 utils stats graphics grDevices
methods
[9] base
other attached packages:
[1] ggplot2_0.8.3 reshape_0.8.3 proto_0.3-8 mvbutils_2.2.0
[5] ChemoSpec_1.2 lattice_0.17-25 mvoutlier_1.4 plyr_0.1.8
[9] RColorBrewer_1.0-2 chemometrics_0.4 som_0.3-4
robustbase_0.4-5
[13] rpart_3.1-45 pls_2.1-0 pcaPP_1.7 mvtnorm_0.9-7
[17] nnet_7.2-48 mclust_3.2 MASS_7.2-48 lars_0.9-7
[21] e1071_1.5-19 class_7.2-48
Here is one way:> fo <- y ~ f1 * f2 > one.x <- lapply(all.vars(fo[[3]]), function(x) { fo[[3]] <- as.name(x); fo }) > one.x[[1]] y ~ f1 [[2]] y ~ f2 On Wed, Oct 21, 2009 at 11:29 AM, Bryan Hanson <hanson at depauw.edu> wrote:> Hello All.. > > Please consider the following: > > y <- rnorm(20, mean = 10) > f1 <- as.factor(rep(c("A", "B", "B", "A"), 5)) > f2 <- as.factor(rep(c("C", "D"), 10)) > testdata <- data.frame(y, f1, f2) > > testFunc <- function(formula, data, ...) { > # ? ?mf <- model.frame(formula, data) > ? ?kw.res <- kruskal.test(formula, data) > ? ?} > > res <- testFunc(y ~ f1 * f2, testdata) > > Which works perfectly well. ?Now, I would like to do some other tests which > allow only one factor, for instance wilcox.test. ?So, I would like to modify > my formula so that it reads y ~ f1 or y ~ f2, then conduct more tests. ?What > is the smart way to subset or modify an existing symbolic formula? ?I have > been staring at the attributes of res and mf (commented out currently) but > these are nested in a way I find difficult to extract. ??terms, > ?terms.formula, ?formula etc all discuss these attributes but as I said, it > seems a bit impenetrable. ?RSiteSearch("symbolic formula") returns too many > answers. ?No doubt I am missing the obvious, as this is my first foray into > using formulas in my own functions. > > TIA, Bryan > ************* > Bryan Hanson > Acting Chair > Professor of Chemistry & Biochemistry > DePauw University, Greencastle IN USA > > >> sessionInfo() > R version 2.10.0 RC (2009-10-19 r50172) > i386-apple-darwin8.11.1 > > 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 ? ? ?utils ? ? stats ? ? graphics ?grDevices > methods > [9] base > > other attached packages: > ?[1] ggplot2_0.8.3 ? ? ?reshape_0.8.3 ? ? ?proto_0.3-8 ? ? ? ?mvbutils_2.2.0 > ?[5] ChemoSpec_1.2 ? ? ?lattice_0.17-25 ? ?mvoutlier_1.4 ? ? ?plyr_0.1.8 > ?[9] RColorBrewer_1.0-2 chemometrics_0.4 ? som_0.3-4 > robustbase_0.4-5 > [13] rpart_3.1-45 ? ? ? pls_2.1-0 ? ? ? ? ?pcaPP_1.7 ? ? ? ? ?mvtnorm_0.9-7 > [17] nnet_7.2-48 ? ? ? ?mclust_3.2 ? ? ? ? MASS_7.2-48 ? ? ? ?lars_0.9-7 > [21] e1071_1.5-19 ? ? ? class_7.2-48 > > ______________________________________________ > R-help at 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. >
Thanks Gabor, you taught me two useful things: all.vars, and the fact that a formula object always has length 3. Problem solved. Bryan On 10/21/09 11:57 AM, "Gabor Grothendieck" <ggrothendieck at gmail.com> wrote:> Here is one way: > >> fo <- y ~ f1 * f2 >> one.x <- lapply(all.vars(fo[[3]]), function(x) { fo[[3]] <- as.name(x); fo }) >> one.x > [[1]] > y ~ f1 > > [[2]] > y ~ f2 > > > On Wed, Oct 21, 2009 at 11:29 AM, Bryan Hanson <hanson at depauw.edu> wrote: >> Hello All.. >> >> Please consider the following: >> >> y <- rnorm(20, mean = 10) >> f1 <- as.factor(rep(c("A", "B", "B", "A"), 5)) >> f2 <- as.factor(rep(c("C", "D"), 10)) >> testdata <- data.frame(y, f1, f2) >> >> testFunc <- function(formula, data, ...) { >> # ? ?mf <- model.frame(formula, data) >> ? ?kw.res <- kruskal.test(formula, data) >> ? ?} >> >> res <- testFunc(y ~ f1 * f2, testdata) >> >> Which works perfectly well. ?Now, I would like to do some other tests which >> allow only one factor, for instance wilcox.test. ?So, I would like to modify >> my formula so that it reads y ~ f1 or y ~ f2, then conduct more tests. ?What >> is the smart way to subset or modify an existing symbolic formula? ?I have >> been staring at the attributes of res and mf (commented out currently) but >> these are nested in a way I find difficult to extract. ??terms, >> ?terms.formula, ?formula etc all discuss these attributes but as I said, it >> seems a bit impenetrable. ?RSiteSearch("symbolic formula") returns too many >> answers. ?No doubt I am missing the obvious, as this is my first foray into >> using formulas in my own functions. >> >> TIA, Bryan >> ************* >> Bryan Hanson >> Acting Chair >> Professor of Chemistry & Biochemistry >> DePauw University, Greencastle IN USA >> >> >>> sessionInfo() >> R version 2.10.0 RC (2009-10-19 r50172) >> i386-apple-darwin8.11.1 >> >> 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 ? ? ?utils ? ? stats ? ? graphics ?grDevices >> methods >> [9] base >> >> other attached packages: >> ?[1] ggplot2_0.8.3 ? ? ?reshape_0.8.3 ? ? ?proto_0.3-8 ? ? ? ?mvbutils_22.0 >> ?[5] ChemoSpec_1.2 ? ? ?lattice_0.17-25 ? ?mvoutlier_1.4 ? ? ?plyr_0.1.8 >> ?[9] RColorBrewer_1.0-2 chemometrics_0.4 ? som_0.3-4 >> robustbase_0.4-5 >> [13] rpart_3.1-45 ? ? ? pls_2.1-0 ? ? ? ? ?pcaPP_1.7 ? ? ? ? ?mvtnorm_0.9-7 >> [17] nnet_7.2-48 ? ? ? ?mclust_3.2 ? ? ? ? MASS_7.2-48 ? ? ? ?lars_0.9-7 >> [21] e1071_1.5-19 ? ? ? class_7.2-48 >> >> ______________________________________________ >> R-help at 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. >>