Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

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

Re: st: MIXLOGIT: marginal effects

From   Clive Nicholas <>
Subject   Re: st: MIXLOGIT: marginal effects
Date   Wed, 8 Feb 2012 07:53:28 +0000

Nick Cox replied:

> I love logits too and I am not especially an advocate of the linear
> probability model, but its defence seems simple. It is a model people
> might want to consider if it fits fairly well over the range of the
> data, not least because statistical people find probability a useful
> scale to think on. This can happen when the response proportion
> doesn't vary very much. Naturally I agree that a logit model might
> work as well or better in this circumstance.


Maarten Buis replied:


... [I]n case of a fully saturated model, it is really a matter of
whether you want your parameters in terms of differences in
probabilities or ratios of odds.

In models that do not include all interactions or where you add a
continuous explanatory variable the linear probability model is more
restrictive. However, that does not bother me too much; a model is
after all supposed to be a simplification of reality. You obviously do
want to check that the deviations from linearity or the predictions
outside the [0,1] interval are not getting too much out of hand, but I
think there will be many situations where the linear probability model
is perfectly adequate.


These responses are very absorbing, but they really don't change my
proposition that -logit- is to be preferred when doing such analyses;
I'm not exactly reinventing the wheel here.

That said, there was one celebrated use of the linear probability
model in my old field of political science. Michael Lewis-Beck's
(1990) award-winning book used opinion polling data from six countries
(the USA, the UK, France, Spain, Germany and Italy) to investigate the
extent to which there was 'economic voting' amongst individuals. He
found that both retrospective _and_ prospective evaluations of their
national economies exerted statistically significant effects on
people's voting intentions. He also found no statistical evidence for
the theory of political business cycles using aggregate economic and
electoral data from those same six countries.

Lewis-Beck is regarded as one of the world's most renowned exponents
of election forecasting, so I rather suspect he's now too big a
superstar of his field to be slumming it on Statalist. That said, it
does get very cold up in Iowa. :)

Clive Nicholas

[Please DO NOT mail me personally here, but at
<>. Please respond to contributions I make in
a list thread here. Thanks!]

Lewis-Beck M (1990) "Economics and Elections: The Major Western
Democracies", Ann Arbor, MI: The University of Michigan Press
*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index