I was wondering if there is a package that is capable of doing a test to see if missing data is missing completely at random, missing at random, or missing not at random. I understand the littlesMCAR() function was capable of taking a dataframe and testing if missing values were completely random, and the mcar_test() script by Nicholas Tierney is based off of it. However, littlesMcar is now an orphaned package and I have had a strange error message I can't quite parse from mcar_test():
Error: Problem with `mutate()` input `d2`.
x Lapack routine dgesv: system is exactly singular: U[1,1] = 0
ℹ Input `d2` is `purrr::pmap_dbl(...)`.
ℹ The error occurred in group 1: miss_pattern = 1.
I was wondering if there are any alternative packages that I could feed my phenotype data frame to accomplish this analysis of missingness? Barring that, has anyone encountered this error and knows what it means?
Thanks!