[R] r code

[Jim Lemon]

On 01/04/2013 02:03 AM, catalin roibu wrote:
> I have a problem. My data is now in this form:
> plot area   d
> 1     0.01  34.6
> ...................
> 100  0.01  15.7
>
> First, I want to aggregate plots start from a plot in the center of sample
> (10 plots x 10 plots). I want to aggregate  after the model plot1 in the
> center of area (0,01 m2), after increasing plot area with the near plots
> until the entire area (10000 m2). For a good explication I will attach a
> figure.
>
> Second, I want to aggregate plots starting to 1, 1, with 2 , 1 with 2 with
> 3 .....1 with 2 with 3........with 99 with 100.
>   Third>  I want to aggregate the values but choose random>Ex. 75, 9 with
> 83,......etc
> Thank you very much!
>
Hi catalin,
As your figure did not make it through to the list, I will have to do a 
bit of guessing. Your data seem to be 100 samples, each 0.01 M2 in area. 
These add up to 1 M2, so I'm not sure where the 10000 M2 comes from. The 
aggregations are not that difficult, simply specify sets of indices and 
use either the aggregate function you would like in a loop.
Assume that your data are in a data frame "mydata":

for(i in 1:100) cat("mean of 1:",i,mean(mydata[1:i,],na.rm=TRUE),"\n")
for(i in 1:100) {
  ind<-sample(1:100,2)
  cat("mean of",ind[1],"and",ind[2],mean(mydata[ind,],na.rm=TRUE]),"\n")
}

Jim