Multiple merge, better solution?

If you don't mind I've added this example to the R wiki,

http://wiki.r-project.org/rwiki/doku.php?id=tips:data-frames:merge

It would be very nice if a R guru could check that the information I put is
not complete fantasy. Feel free to remove as appropriate.

Best wishes,

baptiste


On 19 Feb 2009, at 11:00, Lauri Nikkinen wrote:

Thanks, both solutions work fine. I tried these solutions to my real
data, and I got an error

Error in match.names(clabs, names(xi)) :
 names do not match previous names

I refined this example data to look more like my real data, this also
produces the same error. Any ideas how to prevent this error?

DF1 <- data.frame(var1 = letters[1:5], a = rnorm(5), b = rnorm(5), c =
rnorm(5))
DF2 <- data.frame(var1 = letters[3:7], a = rnorm(5), b = rnorm(5), c =
rnorm(5))
DF3 <- data.frame(var1 = letters[6:10], a = rnorm(5), b = rnorm(5), c
= rnorm(5))
DF4 <- data.frame(var1 = letters[8:12], a = rnorm(5), b = rnorm(5), c
= rnorm(5))

g <- merge(DF1, DF2, by.x="var1", by.y="var1", all=T)
g <- merge(g, DF3, by.x="var1", by.y="var1", all=T)
merge(g, DF4, by.x="var1", by.y="var1", all=T)

Error in match.names(clabs, names(xi)) :
 names do not match previous names

DF <- DF1
for ( .df in list(DF2,DF3,DF4) ) {

+  DF <-merge(DF,.df,by.x="var1", by.y="var1", all=T)
+ }

Error in match.names(clabs, names(xi)) :
 names do not match previous names

Reduce(function(x, y) merge(x, y, all=T,by.x="var1", by.y="var1"),
list(DF1, DF2, DF3, DF4), accumulate=F)

Error in match.names(clabs, names(xi)) :
 names do not match previous names

- Lauri

2009/2/19 baptiste auguie <ba208 at exeter.ac.uk>:

Hi,


I think Reduce could help you.

DF1 <- data.frame(var1 = letters[1:5], a = rnorm(5))
DF2 <- data.frame(var1 = letters[3:7], b = rnorm(5))
DF3 <- data.frame(var1 = letters[6:10], c = rnorm(5))
DF4 <- data.frame(var1 = letters[8:12], d = rnorm(5))

g <- merge(DF1, DF2, by.x="var1", by.y="var1", all=T)
g <- merge(g, DF3, by.x="var1", by.y="var1", all=T)
g <- merge(g, DF4, by.x="var1", by.y="var1", all=T)

test <- Reduce(function(x, y) merge(x, y, all=T,by.x="var1",
by.y="var1"),
                     list(DF1, DF2, DF3, DF4), accumulate=F)

all.equal(test, g) # TRUE


As a warning, it's the first time I've ever used it myself...


Hope this helps,

baptiste



On 19 Feb 2009, at 10:21, Lauri Nikkinen wrote:

Hello,

My problem is that I would like to merge multiple files with a common
column but merge accepts only two
data.frames to merge. In the real situation, I have 26 different
data.frames with a common column. I can of course use merge many times
(see below) but what would be more sophisticated solution? For loop?
Any ideas?

DF1 <- data.frame(var1 = letters[1:5], a = rnorm(5))
DF2 <- data.frame(var1 = letters[3:7], b = rnorm(5))
DF3 <- data.frame(var1 = letters[6:10], c = rnorm(5))
DF4 <- data.frame(var1 = letters[8:12], d = rnorm(5))

g <- merge(DF1, DF2, by.x="var1", by.y="var1", all=T)
g <- merge(g, DF3, by.x="var1", by.y="var1", all=T)
merge(g, DF4, by.x="var1", by.y="var1", all=T)

Thanks in advance.

-Lauri