Open DarioS opened 3 years ago
This is unavoidable. The only thing you can do is reduce k, find more non-missing values of that variable, or find more samples.
The issue relates to variables which don't have sufficient replication to support the estimation of the number k of factors of unwanted variation. I will add it to the documentation of both functions.
Is there anything the user should do or is it unavoidable? Could it be mentioned in the function's documentation?