residual correlation matrix from spOccupancy

[Claudio]

Hi all,

after fitting JSDM models with spOccupancy (sfJSDM and lfJSDM), I got the best model. Now I would like to obtain the (residual) correlation matrix in order to have infomation on co-occurrence patterns, but I do not know how to do this.

Thanks in advance for your help.

[Jeffrey Doser]

Hi Claudio,

You can derive the residual correlation matrix from the latent factor loadings lambda, which is briefly described in the paper introducing this functionality. Here's some code to do that, assuming that the model output name is "out":

# out is the model object from the factor model run.
N <- nrow(out$y)
n.samples <- nrow(out$beta.samples)
cor.mat.samples <- array(NA, dim = c(n.samples, N, N))
n.factors <- out$q
for (l in 1:n.samples) {
  curr.lambda.mat <- matrix(out$lambda.samples[l, ], N, n.factors)
  cor.mat.samples[l, , ] <- cov2cor(curr.lambda.mat %*% t(curr.lambda.mat))
}
# Summarize cor.mat.samples just like any other set of MCMC samples, e.g.,
apply(cor.mat.samples, c(2, 3), mean)

Jeff

[Claudio]

Great! Thanks.

Now I have just to understand better how it works.

Claudio