Nested Model

[Joshua Smith]

Hi All,

I have to estimate a nested logit model but I'm not able to interpret the parameter. I started with the example in Stata:

webuse restaurant

nlogitgen type = restaurant(fast: Freebirds | MamasPizza, family: CafeEccell | LosNortenos | WingsNmore, fancy: Christophers | MadCows)

nlogittree restaurant type, choice(chosen)

nlogit chosen cost distance rating || type: income kids, base(family) || restaurant:, noconst case(family_id)

The output gives me three coefficients on cost distance rating I cannot interpret: what does each coefficient mean? how is related to the interpretation within nest?

Thank you very much

[Kai Shen Lim]

I'm wondering about this too. Anyone familiar with the interpretation? Googled about and couldn't find any leads. Thanks.

[Alfonso Sánchez-Peñalver]

Hi Joshua and Kai,

cost, distance, and rating are what are called alternative-specific variables, i.e. they vary both across individuals and alternatives. The coefficients on those variables reflect the change in the log-odds of an alternative being selected. An increase in the cost of one of the alternatives in 1 unit, for example, will decrease the log-odd of that alternative being chosen by 0.18. I hope this helps.

[Dyah Pritadrajati]

Hi everyone,

I am experiencing the same problem, did anyone figure out how to get the marginal effects (AME and MEAM) after nlogit command in Stata. Maybe someone can provide an alternative command "by hand" to compute marginal effect after nlogit. I would really appreciate any help.

Thank you,
Dyah