[R] Working with duplicated rows

[David Winsemius]

One of undoubtedly many ways:

# kind of a pain to delete the blank rows. Try to give us full example  
next time?

 > txt<- "long   lat   value
+ 10     20      5
+ 6      2       3
+ 27    -3       9
+ 10     20      10
+ 4     -1       0
+ 6      2       9
+ "
 > DF2 <- read.table(textConnection(txt), header=TRUE)

 > DF3 <- data.frame(with(DF2,list(lon=long, lat=lat,  
value=ave(value,long, lat, FUN=sum))))

 > DF3
   lon lat value
1  10  20    15
2   6   2    12
3  27  -3     9
4  10  20    15
5   4  -1     0
6   6   2    12

If you want the column name to be "long", the fix is obvious.

-- 
David Winsemius
On Jan 12, 2009, at 10:52 AM, Steve Murray wrote:

>
> Dear all,
>
>
>
> I have a dataframe of 3 columns, consisting of 'longitude', 'latitude'
> and a corresponding 'value'. Where identical 'longitude' and  
> 'latitude'
> pairs occur more than once, I want their corresponding 'value' to be
> summed and the 'pair' to only appear once.
>
>
>
> For example:
>
>
>
> long   lat   value
>
> 10     20      5
>
> 6      2       3
>
> 27    -3       9
>
> 10     20      10
>
> 4     -1       0
>
> 6      2       9
>
>
>
>
>
> would be converted to something like:
>
>
>
> long   lat   value
>
>
> 10     20      15
>
>
> 6      2       12
>
>
> 27    -3       9
>
>
> 4     -1       0
>
>
>
>
>
>
> ...as rows 1 and 4, 2 and 6 respectively are matched with respect to
> the 'long' and 'lat' columns. Their values in column 3 are then summed
> and reported as one row in the new dataframe.
>
>
>
> How would I go about coding this in R?
>
>
>
> Many thanks,
>
>
>
> Steve
>
>
> _________________________________________________________________
> Are you a PC?  Upload your PC story and show the world
>
> ______________________________________________
> R-help at 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.