Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: How to correctly estimate nested logit sequentially?

From   Alexander Kihm <>
Subject   st: How to correctly estimate nested logit sequentially?
Date   Thu, 21 Feb 2013 17:53:20 +0100

Dear Statalisters,

I am building a nested logit model with a very large number of
alternatives and different sampling strategies for each nest. For
easier handling I would like to estimate it “old-school” sequentially,
like it was done in [1] and more recently [2]. But I got confused by
the discussions in [3] and [4] about differences in non-normalized and
utility-maximizing nested logit and want to make sure that I
sequentially estimate the latter. I tried to sequentially replicate
many different examples of simulaneously-estimated models (like in [5]
or [6]), be it with the travel dataset (webuse travel) or the
restaurant one, but I don’t manage to compare the results.

Hence I just want to ask:
Can you give me an example code of a nested logit model that you
estimate simultaneously (i.e. by nlogit or nlogitrum) and replicate
the results sequentially (i.e. by separate clogit including the
inclusive values)?

Thank you for your help!


[1] Train, K. (1986). “Qualitative Choice Analysis: Theory,
Econometrics, and an Application to Automobile Demand”, MIT Press

[2] “Follow-on Development of CARBITS: A Response Model for the
California Passenger Vehicle Market,” Final Report (Contract 05-303)
prepared for State of California Air Resources Board, April 30, 2009

[3] Heiss, F. (2002). "Structural Choice Analysis with Nested Logit
Models." Stata Journal, 2(3), 227-252.

[4] Silberhorn, N., Boztug, Y., and Hildebrandt, L. (2006).
"Estimation with the Nested Logit Model: Specification and Software
Particularities." SFB 649 Discussion Paper, SFB649DP2006-017.
Sonderforschungsbereich 649, Humboldt University, Berlin, Germany



*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index