search

VEuPathDB / microbiomeComputations

1 stars 0 forks source link

Handle NA in correlation method #60

Closed d-callan closed 1 year ago

d-callan commented 1 year ago

wont toss out vars w NA, but also wont err if there are no complete cases.

d-callan commented 1 year ago

i dont think its our business to make sure na.or.complete does what it claims.. but i can maybe think of a simple test that tries to catch regressions back to the default behavior of cor.

asizemore commented 1 year ago

Yeah that's what i mean - make sure it does what we mean it to do when we intentionally give it nas. And since i dont think the argument's parameter describes itself that well, the test will help us understand what we did in two months.