Structure and Uncertainty in Discrete Choice Models

Abstract

Social scientists are often confronted with theories in which one or more actors make choices over a discrete set of options. In this article, I generalize a broad class of statistical discrete choice models, with both well-known and new nonstrategic and strategic special cases. I demonstrate how to derive statistical models from theoretical discrete choice models and, in doing so, I address the statistical implications of three sources of uncertainty: agent error, private information about payoffs, and regressor error. For strategic and some nonstrategic choice models, the three types of uncertainty produce different statistical models. In these cases, misspecifying the type of uncertainty leads to biased and inconsistent estimates, and to incorrect inferences based on estimated probabilities.

Weighting Parties and Coalitions: How Coalition Signals Influence Voting Behavior

“…The idea to derive such a sequential choice model is strongly influenced by Signorino (2003). Our adaptation, however, is tailored to analyze sequential choice decision in survey experiments.…”

“…13 10. It is well-known that iid violations can be simply produced by misspecified systematic components (Signorino 2003). A preferable strategy in many applied settings is therefore to account for potential sources of iid violations in the specification of the systematic component rather than the stochastic component of the model (Dow and Endersby 2004;Glasgow 2001).…”

The Constraining Power of International Treaties: Theory and Methods

“…Heckman selection models 6 have enjoyed a recent burst of popularity in the political science literature (Berinsky 1999;Lemke and Reed 2001;Reed 2000;Timpone 1998;Vreeland 2003), although political methodologists are well aware of the problems with this class of models (Sartori 2003;Signorino 2003). Research has shown that Heckman-style models share several important weaknesses, including their sensitivity to specification, possible problems of collinearity, and heavy reliance on distributional assumptions (Lee 2001;Liao 1995;Sartori;Winship and Mare 1992).…”

Strategic Binary Choice Models with Partial Observability

“…An important empirical implication of strategic interactions is that observable data are characterized by non-ignorable missingness, i.e. data are missing not at random (MNAR) (Signorino, 1999(Signorino, , 2003Signorino and Yilmaz, 2003;Nieman, 2015). Consider, for example, a model of the effectiveness of extended deterrence, i.e.…”