I am computing pccs in R using a dataset "X" of dimension mxp (m>>p). Notably, all goes well when "X" has no missing values (no "NA" in dataset). So far I have tried commands "pcor" and "pcor.test". The problem is that the data I am working with has some missing values ("NA") and that creates a problem with computing partial correlations. Any tips, anyone?

More Jimmy Omony's questions See All
Similar questions and discussions