Here is the WinBUGS code
model {
for(i in 1:N) {m[i] <- 1/n[ind[i]] }
cumsum[1] <- 0
for(i in 2:(N+1)) {cumsum[i] <- sum(num[1:(i-1)]) }
for(k in 1:sumNumNeigh) {
for(i in 1:N) {
# #pick[k,i] = 1 if cumsum[i] < k <= cumsum[i=1]; otherwise, pick[k,i] = 0
##step(e) 1 if e >= 0; 0 otherwise
pick[k,i]<-step(k-cumsum[i]-epsilon)*step(cumsum[i+1]-k) }
C[k]<-1/ inprod(num[], pick[k,]) }
epsilon <- 0.0001
Uwe Ligges wrote:
If you want that people help to translate *code*, you have to specify
it ...
Uwe Ligges
Jason Gasper wrote:
I am hoping someone can help translate some WinBUGS code into R code.
I would like to use R to create the C[] matrix required for a
car.proper model in WinBUGS, but I am having a difficult time
negotiating the coding. The C matrix provides normalized weights for
each pair of spatial areas. So the WinBUGS example is as follows:
# of the weight matrix with elements Cij. The first J1 elements of
the C[] vector contain the
# weights for the J1 neighbours of area i=1; the (J1+1) to J2
elements of the C[] vector contain
# the weights for the J2 neighbours of area i=2; etc.
# To set up this vector, we need to define a variable cumsum, which
gives the values of J1,
# J2, etc.; we then set up an index matrix pick[,] with N columns
corresponding to the
# i=1,...,N areas, and with the same number of rows as there are
elements in the C[] vector
# (i.e. sumNumNeigh). The elements C[ (cumsum[i]+1):cumsum[i+1] ]
correspond to
# the set of weights Cij associated with area i, and so we set up ith
column of the matrix pick[,]
# to have a 1 in all the rows k for which cumsum[i] < k <=
cumsum[i+1], and 0's elsewhere.
# For example, let N=4 and cumsum=c(0,3,5,6,8), so area i=1 has 3
neighbours, area i=2 has 2
# neighbours, area i=3 has 1 neighbour and area i=4 has 2 neighbours.
The the matrix pick[,] is:
# pick
# 1, 0, 0, 0,
# 1, 0, 0, 0,
# 1, 0, 0, 0,
# 0, 1, 0, 0,
# 0, 1, 0, 0,
# 0, 0, 1, 0,
# 0, 0, 0, 1,
# 0, 0, 0, 1,
#
# We can then use the inner product (inprod(,)) function in WinBUGS
and the kth row of pick to
# select which area corresponds to the kth element in the vector C[];
likewise, we can use inprod(,)
# and the ith column of pick to select the elements of C[] which
correspond to area i.
Basically I want to do this in R to speed things up a little. Has
anyone written a function for this conversion?
______________________________________________
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.
--
Jason Gasper
National Marine Fisheries Service
Alaska Region, Sustainable Fisheries Division
709 W. 9th St. Juneau, Alaska 99801
Juneau, Alaska 99801
Phone 907-586-7237
Fax 907-586-7249
______________________________________________
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.
