Error message while estimating weighted odds ratio

[MK Tan]

Hello,

I am working with hospitalization data and using a multivariable multilevel logistic regression model. Whenever I try to use the sample weights in my model, I encounter an error message stating 'initial values not feasible r(1400)'.

I have a total of 6 million observations in my dataset. I tried with simple linear regression first and started adding the independent variables one by one. Simple regression worked fine for the entire dataset but multilevel model didn't work even for only 300k observations. Please note, I do not receive this error code for running the model without sample weights. I am using Stata SE 14.2 for windows.

Here is what I am receiving for my command:

melogit HOSPITALIZATION i.RACE i.INCOME if _n<300000 [pw=SAMPLEWT] || HOSPITALS:, or

Fitting fixed-effects model:

Iteration 0: log likelihood = -511031.23
Iteration 1: log likelihood = -508117.64
Iteration 2: log likelihood = -508111.86
Iteration 3: log likelihood = -508111.86

Refining starting values:

Grid node 0: log likelihood = -490994.92

Fitting full model:

initial values not feasible
r(1400);

I have been struggling to understand why is this happening and what can I do to solve this issue. Any help would be much appreciated.