Hi:
You could always create a matrix of nCd combinations of size d from n =
length(data) and apply your delete-d jackknife to that. Here's a simple
version for delete-2:
# data vector
> u <- rnorm(10)
> library(combinat)
Attaching package: 'combinat'
The following object(s) are masked from 'package:utils':
combn
# matrix of all combinations of size 2 from 1:10
> combn(10, 2)
[,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10] [,11] [,12] [,13]
[1,] 1 1 1 1 1 1 1 1 1 2 2 2 2
[2,] 2 3 4 5 6 7 8 9 10 3 4 5 6
[,14] [,15] [,16] [,17] [,18] [,19] [,20] [,21] [,22] [,23] [,24] [,25]
[1,] 2 2 2 2 3 3 3 3 3 3 3 4
[2,] 7 8 9 10 4 5 6 7 8 9 10 5
[,26] [,27] [,28] [,29] [,30] [,31] [,32] [,33] [,34] [,35] [,36] [,37]
[1,] 4 4 4 4 4 5 5 5 5 5 6 6
[2,] 6 7 8 9 10 6 7 8 9 10 7 8
[,38] [,39] [,40] [,41] [,42] [,43] [,44] [,45]
[1,] 6 6 7 7 7 8 8 9
[2,] 9 10 8 9 10 9 10 10
# The function below operates on a single column of the above matrix.
# It removes the elements of u corresponding to the indices in each column
# of the matrix above and then computes the average. You can always
# write an external function to pass to apply() for your case.
> apply(combn(10, 2), 2, function(x) mean(u[-x]))
[1] 0.278856785 0.251338791 0.022794855 0.161130397 0.012826789
[6] 0.054276157 0.033911398 0.079530314 -0.039614651 0.364950962
[11] 0.136407026 0.274742569 0.126438960 0.167888328 0.147523569
[16] 0.193142486 0.073997521 0.108889032 0.247224574 0.098920966
[21] 0.140370334 0.120005575 0.165624491 0.046479526 0.018680639
[26] -0.129622970 -0.088173602 -0.108538361 -0.062919445 -0.182064410
[31] 0.008712573 0.050161940 0.029797182 0.075416098 -0.043728867
[36] -0.098141668 -0.118506427 -0.072887511 -0.192032476 -0.077057059
[41] -0.031438143 -0.150583108 -0.051802902 -0.170947867 -0.125328950
It shouldn't be that hard to wrap this up into a function.
HTH,
Dennis
On Wed, Nov 24, 2010 at 5:03 PM, ufuk beyaztas <[email protected]>wrote:
>
> Hi dear all,
> Can aynone help me about delete-d jackknife
> usually normal jackknife code for my data is:
> n <- nrow(data)
> y <- data$y
> z <- data$z
> theta.hat <- mean(y) / mean(z)
> print (theta.hat)
>
> theta.jack <- numeric(n)
> for (i in 1:n)
> theta.jack[i] <- mean(y[-i]) / mean(z[-i])
> bias <- (n - 1) * (mean(theta.jack) - theta.hat)
>
> print(bias)
>
> but how i can apply delete-d jackknife when d=2 or 3.
> Thanks forany idea..
> --
> View this message in context:
> http://r.789695.n4.nabble.com/delete-d-jackknife-tp3058335p3058335.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> [email protected] 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.
>
[[alternative HTML version deleted]]
______________________________________________
[email protected] 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.
