I have a dataset of 547 columns with 643 cases, with a large proportion of missingness, so I am attempt