-project.org] On Behalf Of Zhixin Liu
> Sent: Sunday, December 14, 2008 7:57 PM
> To: r-help@r-project.org
> Subject: [R] how to create duplicated ID in multi-records per
> subject dataset
>
> Hi R helpers,
>
> If I have a dataset looks like:
> ID record
> 1
Try this:
Lines <- "ID;record
1;20
;30
;25
2;26
;15
3;21"
x <- read.table(textConnection(Lines), sep = ";", header = TRUE)
library(zoo)
x$ID <- na.locf(x$ID)
On Sun, Dec 14, 2008 at 11:56 PM, Zhixin Liu wrote:
> Hi R helpers,
>
> If I have a dataset looks like:
> ID record
> 120
> .
andrew has a point which makes my solution wrong. you'd have to change
the factors to numerics and I'm not sure what
would happen when you did that. if you want to send a sample file of
your data, that would be best but andrew's suggestion may
work right off the bat.
On Sun, Dec 14, 2008 at
if the records are in the file dupIDs.txt, then when you read them in,
the IDs become factors. Coercing them to numeric gets them to assign
a unique number to each factor.
So, you could try the following:
dupIDs <- read.table("dupIDs.txt", header = T)
dupIDs$ID2 <- cummax(as.numeric(dupIDs$ID)-1
hi: change your dots to NAs and then use na.locf in the zoo package. i
didn't test it but i think that should work.
DF$ID[DF$ID == .]<-NA
DF$ID<-na.locf(DF$ID)
On Sun, Dec 14, 2008 at 8:56 PM, Zhixin Liu wrote:
Hi R helpers,
If I have a dataset looks like:
ID record 120
.
Hi R helpers,
If I have a dataset looks like:
ID record
120
. 30
. 25
2 26
. 15
3 21
4.
And I want it becomes
ID record
120
130
125
2 26
215
3 21
4.
That
6 matches
Mail list logo
