[Leonardo Guizzetti]

Andrew Musau You are absolutely right. It is better I am more specific. The dependent variable is occupations (at 3-digit level) and I am trying to find the determinants selecting into occupations. I can aggregate the occupations to 1 or 2-digit, however I am trying to keep the occupations as detailed as possible.

A possible approach could be to first look broadly at industry or sector groups to look at flow into these areas, then with secondary models explore the finer selection into occupations within these groups.

[Maarten Buis]

You could look at the work done on micro classes, e.g. https://www.journals.uchicago.edu/do...10.1086/596566 . This can give you ideas on how to estimate such models (log linear models) and how to make the results interpretable.

[Elif Cengen]

Andrew Musau , Stata MP can not handle, neither.

Leonardo Guizzetti My dependent variable with 300 categories is occupation. For the time being I do not prefer to aggregate.

Hong Il Yoo and Maarten Buis Many thanks for the papers, they are both very great and helpful!