Curtis S. Signorino

Strategic Interaction and the Statistical Analysis of International Conflict

Although strategic interaction is at the heart of most international relations theory, it has largely been missing from much empirical analysis in the field. Typical applications of logit and probit to theories of international conflict generally do not capture the structure of the strategic interdependence implied by those theories. I demonstrate how to derive statistical discrete choice models of international conflict that directly incorporate the theorized strategic interaction. I show this for a simple crisis interaction model and then use Monte Carlo analysis to show that logit provides estimates with incorrect substantive interpretations as well as fitted values that can be far from the true values. Finally, I reanalyze a well-known game-theoretic model of war, Bueno de Mesquita and Lalmans (1992) international interaction game. My results indicate that there is at best modest empirical support for their model.

Statistical Backwards Induction: A Simple Method for Estimating Recursive Strategic Models

We present a simple method for estimating regressions based on recursive extensive-form games. Our procedure, which can be implemented in most standard statistical packages, involves sequentially estimating standard logits (or probits) in a manner analogous to backwards induction. We demonstrate that the technique produces consistent parameter estimates and show how to calculate consistent standard errors. To illustrate the method, we replicate Leblang’s (2003) study of speculative attacks by financial markets and government responses to these attacks.

Discriminating Methods: Tests for Non-nested Discrete Choice Models

We consider the problem of choosing between rival statistical models that are non-nested in terms of their functional forms. We assess the ability of two tests, one parametric and one distribution-free, to discriminate between such models. Our Monte Carlo simulations demonstrate that both tests are, to varying degrees, able to discriminate between strategic and non-strategic discrete choice models. The distribution-free test appears to have greater relative power in small samples.

Simulating International Cooperation under Uncertainty

The repeated prisioners dilemma is representative of a broad range of situations in international security and trade. This article examines the effects of asymmetric noise on the emergence and maintenance of cooperation under such conditions. The results show that positive and negative asymmetric noise have very different effects on strategy performance. For forgiving strategies, positive noise provides a stimulus out of perpetual defection or unsynchronized retaliations, but also opens them to exploitation. For provocable strategies, negative noise triggers unsynchronized retaliations or perpetual defection, although this may be tempered by generosity and contrition. The effects of neutral noise reflects the signature of each asymmetric noise type. Of the strategies examined, contrite tit-for-tat (CTFT) is generally one of the best performers in both homogeneous and heterogeneous systems. Moreover, one generally sees the evolutionary models moving from heterogeneous bilateral interaction to cooperative norms of behavior, often including or even dominated by CTFT.

On Formal Theory and Statistical Methods: A Response to Carrubba, Yuen, and Zorn

The recent article by Carrubba, Yuen, and Zorn (2007) (CYZ) attempts to relate the strategic random utility models in Signorino (1999, 2002, 2003) and in Signorino and Yilmaz (2003) to existing game theory practice and to existing statistical techniques. It contributes to this literature by reminding us that comparative statics analysis can be applied to the equilibria of these models. There are a number of claims in CYZ, however, that require clarification. In particular, the articles primary claim is that comparative statics analysis, in combination with one of three proposed statistical estimators, provides a simpler alternative to methods previously advocated. This claim (or combination of claims) is incorrect. When one examines the procedure CYZ recommends, it is no simpler for substantive researchers than anything previously recommended. Moreover, none of the proposed estimators are new: they are exactly the same methods introduced in Signorino (1999, 2003), in Signorino and Yilmaz (2003), in Signorino, Walker, and Bas (2002), and in Bas, Signorino, and Walker (2007).

Knowing one’s future preferences: A correlated agent model with Bayesian updating

We generalize two classes of statistical sequential incomplete information games: (1) those resembling typical signaling games, in which a single agent represents each player, allowing for information to be revealed about future play; and (2) those in which each player is represented by a set of independent agents, where moves do not reveal private information. The generalized model we develop, the Correlated Agent Model, relies on a parameter, ρ, which denotes the correlation between two agents’ private information, i.e. the extent to which a player knows the future private component of her preferences. The independent agent and single agent models are special cases, where ρ=0 and ρ=1, respectively. The model also allows 0 < ρ < 1, a class of games which have not yet been considered. We apply the model to crisis bargaining and demonstrate how to estimate ρ, as well as parameters associated with utilities.