Change covariance type to robust in statsmodels.logit

Question

I am trying to change the covariance type from non-robust to robust when doing a logistic regression using stats models in python.

I read the documentation on statsmodels.org but wasn't able to find a solution on how to do it.

If someone on here could help me that would be really awesome.

user_5705174 · Answer

As commented by @Josef, simply specifying the cov_type will work:

model.fit(cov_type="hc0")
model.fit(cov_type="hc1")
model.fit(cov_type="hc2")
model.fit(cov_type="hc3")