When the MSM are fit in the IPTW.fit() statement, it is possible for a PerfectSeparationError to occur. This is more of a concern for the linear-binomial models (as opposed to the other). To deal with this issue, I should update to try-except that fills the RD (or RR) with nan's and throw a warning.
Current behavior would be an error and no other models being fit
When the MSM are fit in the
IPTW.fit()statement, it is possible for aPerfectSeparationErrorto occur. This is more of a concern for the linear-binomial models (as opposed to the other). To deal with this issue, I should update to try-except that fills the RD (or RR) with nan's and throw a warning.Current behavior would be an error and no other models being fit