Thanks, David. I tried converting my rectangular matrix to a network object and then back to a square matrix. This works for a small artificial dataset (e.g. 2x5), but when I try this on my 1790x45 or transposed 45x1790 matrix, R keeps working for a couple of minutes and then crashes completely without having saved any data first. I assume something is wrong with my data? Cheers Philip
David Barron wrote:
> I think you can do this using the network package. Look at the > as.network.matrix and as.matrix.network functions, for example. > >> how can I convert an m x n incidence matrix into an m x m adjacency >> matrix or an n x n adjacency matrix? The current matrix contains binary >> data, hence the new matrix would contain counts of common occurrences.