Of course, sorry... The approach below is indeed a much slicker way to
calculate the probability of the ith variable being the smallest:
#construct a new mv norm distribution, for y = x - x_i
#define y = Bx, where B is a matrix
B - matrix(rep(0,25),nrow=5)
diag(B) - rep(1,5)
B[,i] - -1
B - B[-i,]
On 09 Feb 2014, at 10:56 , Paul Parsons pparsons...@gmail.com wrote:
Many thanks, Peter. Creating a wrapper function for integrand using
Vectorize, and then integrating the wrapper, does indeed solve the problem. I
tried your final suggestion, but the variable x still gets passed into
Many thanks, Peter. Creating a wrapper function for integrand using
Vectorize, and then integrating the wrapper, does indeed solve the
problem. I tried your final suggestion, but the variable x still gets
passed into pmvnorm inside the new mean and variance matrix, leading
to the same
You almost said it yourself: Your integrand doesn't vectorize. The direct
culprit is the following:
If x is a vector, what is lower=c(x,x,x,x)? A vector of length 4*length(x). And
pmvnorm doesn't vectorize so it wouldn't help to have lower= as a matrix (e.g.,
cbind(x,x,x,x)) instead.
A
Hi
I have a multivariate normal distribution in five variables. The
distribution is specified by a vector of means ('means') and a
variance-covariance matrix ('varcov'), both set up as global variables.
I'm trying to figure out the probabilities of each random variable
being the
5 matches
Mail list logo
