Average of individual betas from mixlbeta different from the results in Stata using mixlogit

[Jinhyung Kim]

I recently analyzed my data using mixlogit command and got satisfying results. The code I used is mixlogit choice, rand(Nucl LNG Onshore_wind Solar Sea_solar HFC SMR SMR_dist Dist Nochoice Fee_ln) ln(1) group(GROUP) id(ID). Therefore, to use the individual betas from the results, I used the mixlbeta command to retrieve the individual betas. However, the individual betas for the Fee_ln attribute came out to be too high and very different from the mean beta reported in STATA which was about 1.6, where the average of the retrieved individual beta turned out to be around 105. This has never happened before and it was very frustrating because I do not know why this is happening. What is the problem? Is it a problem in the dataset or is it a problem with the code?