Nerakg wrote on 01/18/2012 04:48:48 AM:
> Dear all,
> I have a question concerning manipulating data of several columns of a
> dataframe at the same time.
> I manage to do it for one column (with the use of the specific name for
this
> column).
> In each columns, I have 60 values. But I should reorganize the values
> (because I created this as an output before and I want to compare it
with an
> other dataset). I want that the value on row 2 becomes the value of row
1,
> value 3 value 2 and so on. The first value would be NA.
Try this.
# Combine one row of NAs, with the original data frame, minus the last row
results <- rbind(rep(NA, dim(new)[2]), new[-dim(new)[1], ])
new[c(1:3, 58:60), 1:6]
depth_1 depth_1.5 depth_2 depth_2.5 depth_3 depth_3.5
1 1 2 3 1 2 3
2 2 3 4 2 3 4
3 3 4 5 3 4 5
58 58 59 60 58 59 60
59 59 60 61 59 60 61
60 60 61 62 60 61 62
results[c(1:3, 58:60), 1:6]
depth_1 depth_1.5 depth_2 depth_2.5 depth_3 depth_3.5
1 NA NA NA NA NA NA
2 1 2 3 1 2 3
3 2 3 4 2 3 4
58 57 58 59 57 58 59
59 58 59 60 58 59 60
60 59 60 61 59 60 61
Jean
> If I would do this for 1 column (with the name depth_1), I would do it
like
> this:
>
> for (t in 2:60)
> {
> results$depth[t]<-new$depth_1[t-1]
> }
>
>
> # But in my dataset I have 91 columns and I would like to find a way not
> having to write this for every column?
> # I cannot give my dataset where I?m working on so I created one just
for
> trying it out and to provide a reproducible example. I created a data
frame
> ?new? with 26 columns and 60 rows. I named the columns all in a similar
way
> using ?C <- seq(1,13.5,0.5)?. That means that all my column names are
> structured in the same way: depth_1 ; depth_1.5, depth_2; depth_2.5 and
so
> on.
>
> C <- seq(1,13.5,0.5)
>
> a<-c(1:60)
> b<-c(2:61)
> c<-c(3:62)
> d<-c(1:60)
> e<-c(2:61)
> f<-c(3:62)
> g<-c(1:60)
> h<-c(2:61)
> i<-c(3:62)
> j<-c(1:60)
> k<-c(2:61)
> l<-c(3:62)
> m<-c(1:60)
> n<-c(2:61)
> o<-c(3:62)
> p<-c(1:60)
> q<-c(2:61)
> r<-c(3:62)
> s<-c(1:60)
> t<-c(2:61)
> u<-c(3:62)
> v<-c(1:60)
> w<-c(2:61)
> x<-c(3:62)
> y<-c(1:60)
> z<-c(2:61)
>
> new<-data.frame(a,b,c,d,e,f,g,h,i,j,k,l,m,n,o,p,q,r,s,t,u,v,w,x,y,z)
> names(new)<-c(paste('depth_',C,sep=''))
>
> #I don't know if I may create a now dataframe that's empty and then
binding
> my results to it? Or should I create a dataframe with already one column
> with the amount of rows that I will have in my new dataset? #results <-
> data.frame() is what I would use to create an empty
> dataframe but that gives some problems when I want to add other columns
> because the length is different.
> results <- data.frame(time=1:60)
> ### I use n in 2:27 because this needs to be done on column 2 till
column
> 27, not on column 1
> for (n in 2:27)
> {
>
> # I don?t know if I should indicate that there should be created an new
> variable with 60 rows? I saw people doing it in a script, but it didn?t
seem
> necessary in an other script that was similar to it and I made myself?
> # results$newdepth<-(1:60)
> #next line I also don?t know if I should give it. If I did this whole
thing
> for only 1 column, the firs row was NA without asking for it.
>
> results$newdepth[1,n]<-NA
>
> for (t in 2:60)
> {
> results$newdepth[t,n] <- new[t-1,n]
> }
> results<- cbind(results, results$newdepth)
> }
>
> names(results) <- c(names(results)[c(1)],paste('newdepth_',C,sep=''))
> #for example:
> results$newdepth_1.5
>
> I hope someone can help me with this? I hope I gave enough information?
I
> think there should be an easier manner but I really have no other idea.
> I?m also wondering if it?s possible not using column numbers in a
function,
> but the name if those are structured in the same way? Eg.
depth_?followed by
> a number? If you want for example to do something with the columns
depth_1,
> depth_1.5, depth_2, depth_2.5, depth_3 ? belonging to a dataframe with
also
> other columns not related with the name depth (Like here, there is also
the
> column time). I?m hoping there is something useful for that.
>
> Many thanks,
> Nerak
[[alternative HTML version deleted]]
______________________________________________
R-help@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.
