R help - Mar 2012 - Accessing variable's name as an attribute?

Sorry - I suspect this is a very basic query.

I have a data frame structured "flat" (i.e. separate observations for
a given subject ID on separate lines). Using the nlme library I create a
groupedData object suitable for further analysis thus:

dataset <- groupedData(VARIABLE ~ TIME|ID, data=sfa)

The dataset has a large number of VARIABLES of interest so I want to make use of
loops where possible to build groupedData objects from each of the relevant
columns along the lines of:

sfa <- na.omit(read.csv("SFAFinalData.csv", header = TRUE))

factorsList <- colnames(sfa)
noFactors <- length(factorsList)
for (i in seq(4,noFactors)) {

	columnName <- factorsList[i]
	dataset <- groupedData(sfa[,columnName] ~ TIME|ID, data=sfa, labels =
list("Weeks post injury", columnName))

	}


Of course at present this simply results in sequential overwriting of dataset
with the latest groupedData object. What I want to do with each iteration of the
loop I want to create another uniquely named groupedData object - I just
don't know how to assign a unique variable name.

names(dataset) is not the right attribute (this alters the names of the internal
column names) - can I set a variable name attribute directly?

thanks

Rob

Hello,
>
> What I want to do with each iteration of the loop I want to create another
> uniquely named
> groupedData object - I just don't know how to assign a unique variable
> name. 
> 
The best way should be to create a list of groupedData objects. Along the
lines of


library(nlme)

n <- 100
df1 <- data.frame(A=sample(LETTERS[1:5], n, T),
		B=sample(letters[1:5], n, T),
		C1=rnorm(n),
		C2=rnorm(n))

factorsList <- colnames(df1)
# First initialize to object of class 'list'
dataset <- list()
for(i in 3:4){
	# If you want another name for each groupedData object:
	# grname <- paste("gr", factorsList[i], sep="")
	grname <- factorsList[i]
	dataset[[ grname ]] <- groupedData(df1[, i] ~ A | B, data=df1)
}

str(dataset)

# Now use their unique names with the operator '$' or with '[['
dataset$C1
Grouped Data: df1[, i] ~ A | B
    A B           C1           C2
1   C e  0.001311565  2.237548657
2   B d  1.393674594  0.507935464
3   C c -0.821700986 -2.118189680
4   C d  0.314210639 -0.771531899
[... etc ...]
99  C e -0.435389239  1.024320656
100 D e  0.919110919  0.667429797

Hope this helps,

Rui Barradas


--
View this message in context:
http://r.789695.n4.nabble.com/Accessing-variable-s-name-as-an-attribute-tp4475564p4475793.html
Sent from the R help mailing list archive at Nabble.com.