Problem subsetting: undefined columns

Dear R-users,
-I am new to R, and I am struggling with the following problem.

-I am repeating the following  operations hundreds of times, within a loop:
I want to subset a data frame by columns. I am interested in the columns
names that are given by the rows of another data frame that was built in
parallel. The solution I have so far works well as long as the elements of
the second data frame are included in the column names of the first data
frame but if an element from the second object is not a column name of the
first one, then it bugs.

-More concretely, I have the following data frames d and v:
yyyymmdd<-c("19720601", "19720602", "19720605")
sret.10006<-c(1,2,3)
sret.10014<-c(5,9,7)
sret.10065<-c(10,2,11)


d<- data.frame(yyyymmdd=yyyymmdd, sret.10006=sret.10006,
sret.10014=sret.10014, sret.10065=sret.10065)

v<- data.frame(V1="sret.10006", V2="sret.10090")
v<- sapply(v, function(x) levels(x)[x])

-I want to do the following subsetting:
sub<- subset(d, select=c(v))


and I get the following error message:
Error in `[.data.frame`(x, r, vars, drop = drop) :
  undefined columns selected



Any help would be very much appreciated,

Best,
Aurelien

	[[alternative HTML version deleted]]

Dear R-users,
-I am new to R, and I am struggling with the following problem.

-I am repeating the following  operations hundreds of times, within a loop:
I want to subset a data frame by columns. I am interested in the columns
names that are given by the rows of another data frame that was built in
parallel. The solution I have so far works well as long as the elements of
the second data frame are included in the column names of the first data
frame but if an element from the second object is not a column name of the
first one, then it bugs.


-More concretely, I have the following data frames d and v:
yyyymmdd<-c("19720601", "19720602", "19720605")
sret.10006<-c(1,2,3)
sret.10014<-c(5,9,7)
sret.10065<-c(10,2,11)


d<- data.frame(yyyymmdd=yyyymmdd, sret.10006=sret.10006,
sret.10014=sret.10014, sret.10065=sret.10065)

v<- data.frame(V1="sret.10006", V2="sret.10090")
v<- sapply(v, function(x) levels(x)[x])

-I want to do the following subsetting:
sub<- subset(d, select=c(v))


and I get the following error message:
Error in `[.data.frame`(x, r, vars, drop = drop) :
 undefined columns selected



Any help would be very much appreciated,

Best,
Aurelien

   [[alternative HTML version deleted]]

Hi Paul and Jim,
Thanks for your messages.

I just wanted R to give me the columns of my data frame d, whose names
appear in v. I do not care about the names of v that are not in d. In
addition, every time, there will be at least one element of v that has a
corresponding column in d, for sure, so I know there is at least one match
between the 2.

Initially, I  tried something in the spirit:
sub<- subset(d, colnames(d) %in% v)

but I could not make it work properly.


Best,
Aurelien

2011/12/2 Paul Hiemstra <paul.hiemstra at knmi.nl>

On 12/02/2011 07:20 AM, Aur?lien PHILIPPOT wrote:

Dear R-users,
-I am new to R, and I am struggling with the following problem.

-I am repeating the following  operations hundreds of times, within a

loop:

I want to subset a data frame by columns. I am interested in the columns
names that are given by the rows of another data frame that was built in
parallel. The solution I have so far works well as long as the elements

of

the second data frame are included in the column names of the first data
frame but if an element from the second object is not a column name of

the

first one, then it bugs.

Hi Aurelien,

I would call this a feature, not a bug. I think R does what it should
do, you request a non-existent column and it throws an error. What kind
of behavior are you looking for instead of this error?

regards,
Paul

-More concretely, I have the following data frames d and v:
yyyymmdd<-c("19720601", "19720602", "19720605")
sret.10006<-c(1,2,3)
sret.10014<-c(5,9,7)
sret.10065<-c(10,2,11)


d<- data.frame(yyyymmdd=yyyymmdd, sret.10006=sret.10006,
sret.10014=sret.10014, sret.10065=sret.10065)

v<- data.frame(V1="sret.10006", V2="sret.10090")
v<- sapply(v, function(x) levels(x)[x])

-I want to do the following subsetting:
sub<- subset(d, select=c(v))


and I get the following error message:
Error in `[.data.frame`(x, r, vars, drop = drop) :
 undefined columns selected



Any help would be very much appreciated,

Best,
Aurelien

     [[alternative HTML version deleted]]

   [[alternative HTML version deleted]]