1 Automated Syllabus of Game Theory Papers
Built by Rex W. Douglass @RexDouglass ; Github ; LinkedIn
Papers curated by hand, summaries and taxonomy written by LLMs.
2 Comparative statics
2.1 Monotone Comparative Statics
Carefully consider the definitions of your fundamental concepts and ensure they align with established terminology, as well as thoroughly check your proofs to avoid missing critical steps. (Kotani et al. 2023)
Consider using iterative fixed-point comparative statics to analyze the effects of parameter changes in economic systems, particularly those involving strategic complementarities, as this approach allows for unstable equilibria, divergent learning processes, and unordered perturbations. (Balbus et al. 2022)
Consider using the i-directional set order, which is a reformulation of the Ci-flexible set order, to study monotone comparative statics in situations where traditional lattice-based approaches are not applicable due to non-standard constraint sets. (Barthel and Sabarwal 2017)
Utilize a two-step approach to conduct a likelihood-ratio test for order restrictions on the conditional quantiles of Y given X, first constructing nonparametric estimators for the conditional quantiles and then developing a test based on your asymptotic distributions, while accounting for the presence of numerous nuisance parameters due to unknown equilibrium selection probabilities. (“Structural Econometric Models” 2013)
Employ a novel likelihood-ratio test for order restrictions on the conditional quantiles of Y given X, rather than relying solely on traditional methods such as OLS regression, when working with multiple equilibrium models where the MCS property holds. (Echenique and Komunjer 2013)
Utilize ordinal conditions rather than relying solely on traditional assumptions like smoothness, linearity, or convexity when conducting comparative statics analyses, as demonstrated through the development of a theory and methods for comparative statics analysis based on ordinal conditions alone. (Mitsilegas 2012)
Consider leveraging the power of the monotonicity theorem of Topkis (1978) to analyze complex systems involving multiple interacting components, especially when those interactions exhibit complementarities and the system can be represented as a lattice. (Arkolakis 2010)
Consider relaxing the assumption of antisymmetry in binary relations when conducting comparative statics analysis, allowing for a more flexible framework that can handle constrained optimization problems with nonlinear constraints. (Shirai 2010)
Consider using lattice programming techniques and flexible set orders to study comparative statics problems in constrained optimization, particularly when dealing with non-smooth, non-interior, non-convex, or non-unique solutions. (Quah 2007)
Consider reparameterizing your optimization problems to achieve monotone comparative statics, which can be done by identifying a vector field indicating the direction of monotonicity in the parameter space and then transforming the problem using this vector field. (Strulovici and Weber 2007)
Utilize the monotone comparative statics approach to generate empirical predictions that are robust to model misspecification and can be tested using ordinal information and nonparametric methods, thereby avoiding reliance on potentially unfounded technical assumptions. (Ashworth and Mesquita 2005)
Consider using log-supermodularity as a tool for deriving comparative statics predictions, particularly when dealing with uncertain environments, as it provides a powerful framework for understanding how changes in parameters affect optimal choices. (Athey 2002)
Consider using log-supermodularity as a tool for deriving comparative statics predictions, particularly when dealing with uncertain environments, as it provides a strong condition for monotonicity that is preserved under integration. (Athey 2002)
Utilize the concept of “monotone comparative statics” to analyze how changes in exogenous parameters impact endogenous outcomes in models, particularly when dealing with optimization problems. This approach provides ordinal answers regarding whether an increase in a parameter leads to an increase or decrease in the decision variable, while requiring fewer assumptions compared to alternative methods like the implicit function theorem. (NA?)
2.2 Comparative Statics
Consider using the provided sufficient condition and algorithm to identify a minimum threshold parameter value at which every old equilibrium becomes strictly smaller than every new equilibrium, allowing for stronger conclusions about comparative statics in games of strategic complementarity. (Chambers, Echenique, and Saito 2016)
Leverage the power of “robust” comparative statics to analyze large dynamic economies, which provides generalizable insights into how stationary equilibria respond to exogenous shocks and changes in the distribution of idiosyncratic shocks, without requiring extensive simulation or numerical analysis. (Acemoglu and Jensen 2015)
Carefully consider the tradeoffs between assumptions about payoff functions and probability distributions when making comparative statics predictions in stochastic optimization problems, as single crossing properties and log-supermodularity are crucial concepts for ensuring the validity of such predictions. (NA?)
2.3 Distributional Comparative Statics
- Focus on the concavity or convexity of policy functions when studying distributional comparative statics, as long as they understand the conditions under which these policy functions will be concave or convex. (Jensen 2017)
3 Game theory
3.1 Comparative Statics
- Carefully distinguish between “strongly competitive” and “weakly competitive” games of incomplete information, as the comparative statics predictions differ significantly between them. Specifically, in strongly competitive games, even weak types are motivated to play more aggressively in response to a stochastically higher distribution of types, whereas in weakly competitive games, weak types are discouraged and compete less aggressively in a more competitive environment. (Vives 2005)
3.2 Game Theory
Carefully control and standardize experimental procedures across sites to ensure uniformity and minimize potential confounding variables when conducting cross-cultural studies. (Henrich et al. 2006)
Consider conducting cross-cultural studies using diverse samples from various economic and cultural backgrounds to capture the variability in human behavior and challenge assumptions of universality in economic models. (Henrich et al. 2001)
3.3 Single Crossing Property
Look for conditions under which an additively separable objective function satisfies the Milgrom-Shannon single crossing property, specifically when one component allows a monotone concave transformation with increasing differences and is nondecreasing in the parameter variable, and the other component exhibits increasing differences and is nonincreasing in the choice variable. (“Monotone Comparative Statics with Separable Objective Functions” 2010)
Utilize a stronger differential version of the single crossing property and argue from first-order conditions to establish strict monotonicity in comparative statics analyses. (Edlin and Shannon 1998)
3.4 Supermodular Game
Consider using supermodular games, which are based on monotone comparative statics and supermodular optimization, to analyze non-cooperative situations where an increase in one players strategy leads to increases in other players strategies. This approach can provide insights into the existence and stability of equilibria, and avoid assumptions associated with traditional comparative statics methods. (“Game Theory” 2010)
Consider using monotone comparative statics methods instead of the implicit-function theorem for comparative static analysis, as they offer greater flexibility and ease of use while requiring fewer assumptions, such as differentiability, concavity, and convexity. (Tremblay and Tremblay 2010)
3.5 Aggregative Game
- Leverage the aggregative structure of games, where each players payoff depends on her own actions and some aggregate of all players actions, to derive robust and general comparative static results under considerably weaker conditions than traditional approaches. (Acemoglu and Jensen 2009)
3.6 Cheap Talk
- Consider incorporating anonymous messaging in public goods games to encourage cooperation, as the study found that anonymous exhortations significantly increased contributions compared to groups without messaging. (Baum et al. 2012)
3.7 Contest Theory
- Consider applying lattice-theoretic tools to analyze games with non-monotonic best-replies, even if they were initially developed for games of strategic complements/substitutes, because these methods can still provide valuable insights about comparative statics and equilibrium properties. (Gama and Rietzke 2019)
3.8 Dynamic Games
- Carefully address unobserved heterogeneity in dynamic games through methods such as fixed effect conditional likelihood or functional differencing, as failing to do so can lead to biased estimates of structural parameters that capture both dynamic state dependence and strategic interactions among players. (Dobronyi, Gu, and Kim 2021)
3.9 Market Games
- Consider using the theory of supermodular optimization and games to analyze the comparative statics of strategic market games, particularly when the model involves only one good on each side of the market, as this allows for a straightforward analysis based on familiar notions from classical microeconomic theory such as normality and gross substitutes of goods. (Lahiri 2011)
3.10 Persuasion Game
- Examine the “crater property” of interim payoffs to determine if coarse-convexity shifts lead to more informative signals being chosen by senders, regardless of the prior. (Curello and Sinander 2022)
3.11 Quasi-supermodularity
- Carefully distinguish between different types of monotonicity (i.e., type A and type B problems) and consider the appropriate single crossing conditions and quasisupermodularity assumptions required for each type to ensure valid inferences. (Kukushkin 2011)
3.12 Shapley Value
- Carefully consider the implications of using either conditional or interventional value functions when interpreting Shapley values as feature importance measures, as each approach introduces distinct trade-offs and potential biases. (Kumar et al. 2020)
3.13 Substitutes
- Look for conditions under which the indirect strategic substitute effect does not dominate the direct parameter effect in games with strategic substitutes, as this ensures the existence of a larger equilibrium at a higher parameter value. (Roy and Sabarwal 2010)