Using WinBUGS from R: A Multi-Way Array Problem

Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.snow at imail.org
801.408.8111


> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-
> project.org] On Behalf Of Khurram Nadeem
> Sent: Wednesday, June 03, 2009 11:37 PM
> To: r-help at r-project.org
> Subject: [R] Using WinBUGS from R: A Multi-Way Array Problem
> 
> Please suggest a way out to the following problem.
> 
> I have a T by n data matrix (say Y) where coulmns are time series of
> length
> T.
> To do some analysis in WinBUGS I need to construct my data as follows.
> 
> yy<-rep(Y,k) ## this will be a vector
> Yk<-array(yy,dim=c(T,n,k)) ## data array
> 
> Here the definition of dim indices is
> 
> first index: T rows
> second index: n columns
> third index: for kth T by n array
> 
> EXAMPLE
> T=3
> n=2
> k=2
> Y<-matrix(c(1,2,3,4,5,6), ncol=n) ## my data matrix
> yy<-rep(Y,k)
> Yk<-array(yy,dim=c(T,n,k))
> Yk  # this produces the following R output.
> 
> , , 1
>      [,1] [,2]
> [1,]    1    4
> [2,]    2    5
> [3,]    3    6
> , , 2
>      [,1] [,2]
> [1,]    1    4
> [2,]    2    5
> [3,]    3    6
> 
> That is, I have copied the orignal data k=2 times. WinBUGS will be
> supplied
> the following data
> 
> d<-list(Yk=Yk, T=T,n=n,k=k).
> 
> Now in WinBUGS I have to define a multivariate stochastic node as as
> follows
> 
> e[i,1:n,kk]~dmnorm(   ,   ) ## i= 1,2,...,T ; kk = 1,2,...,k
> 
> But Winbugs accepts only something like
> 
> e[ , , 1:n]~dmnorm(   ,   ).
> 
> That is "1:n" has to be given at the leftmost position.
> 
> This means that I need to change the definition of dim indices in R as
> follows.
> 
> first index: for kth T by n array
> second index: T rows
> third index: n columns.
> 
> Specifically, I want the above output as
> 
>  1, ,
>      [,1] [,2]
> [1,]    1    4
> [2,]    2    5
> [3,]    3    6
> 
> 2, ,
>      [,1] [,2]
> [1,]    1    4
> [2,]    2    5
> [3,]    3    6
> 
> I would appriciate any help in this regard.
> 
> 
> Khurram Nadeem
> PhD Student
> Department of Math. & Stat. Sciences
> University of Alberta
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> 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.