Computational Modeling Essay ⋆ Political Science Essay Examples ⋆ EssayEmpire

This example Computational Modeling Essay is published for educational and informational purposes only. If you need a custom essay or research paper on this topic please use our writing services. EssayEmpire.com offers reliable custom essay writing services that can help you to receive high grades and impress your professors with the quality of each essay or research paper you hand in.

In political economy, computational models simulate the behavior of institutions or individuals, allowing researchers to explore emergent patterns in individual and institutional behavior over time. Computational models complement mathematical models, and also serve as a form of independent theory construction in their own right. This distinguishes computational models from statistical computation for data analysis, because although statistical models may involve simulation of mathematical functions, they use simulation to approximate known statistical models that are difficult or impossible to analyze analytically.

Although some scholars used computers to model political behavior in the early 1960s, many of the fundamental ideas now used in computational political economy appeared much later. In particular, Thomas Schelling’s pioneering work in 1978, Micromotives and Macrobehavior, which created micro simulations of individuals without computers, showed dramatically how complex and unexpected patterns of behavior could emerge from individuals acting with simple motives and simple rules of individual behavior. This directly influenced the first major work of computational political economy in 1981, Robert Axelrod’s “The Evolution of Cooperation,” illustrating how cooperative behavior can emerge from self-interested agents operating with simple heuristics.

Modern computational models that describe the behavior of individual actors are sometimes known as agent-based simulations. In most modern agent-based simulations, local interactions are important: Individuals are modeled as acting on locally available information and as interacting with other local agents. Also, in typical agent-based simulations, individuals are modeled as being bounded rational: Agents use heuristics to make decisions rather than acting optimally (in the game-theoretic sense). Moreover, the institutional environment in which individuals act is characterized as both stochastic and dynamic—evolving with, or coevolving in reaction to, the behavior of individual agents.

Computational models do not require individuals as the modeling unit. For example, models of international conflict, in which nations are the fundamental actors, date back to the early 1950s.Although used less frequently in political economy, institutional-level models are common in macroeconomics and finance.

Although initially opposed by formal theorists as too imprecise, and by qualitative theorists as too impoverished, computational models have gained a share of acceptance in the last decade. As a complement to mathematical theory, computational models are most often advocated as a way to generate both examples and counterexamples with which to probe the robustness of the mathematical model for changes in assumptions. Computational models may also be used as a constructive form of theory building, independent of a formal mathematical model, as the basis for making predictions and for generating qualitative insights. As such, they are often justified as a middle ground between purely mathematical formal models and purely textual qualitative models. Because computational models are far easier to construct than formal mathematical models, the researcher can use them to obtain, in the happiest of circumstances, the precision of a formal model with the realism of a qualitative model.

Proponents of computational models argue further that dynamic computational models are better-fitted models for studying dynamic patterns than standard mathematical equilibrium models. (Using equilibrium models to study dynamic behavior is sometimes likened to trying to understand Niagara Falls by staring into a collection bucket.) Still, even ardent proponents emphasize the need for caution in model building and interpretation. As in other forms of model building, seemingly innocuous assumptions may sometimes yield striking different patterns of outcomes. Thus, all models should be built with care, and researchers should actively seek cases in which competing models yield diverging predictions that may be directly compared.

Bibliography:

Axelrod, Robert. “The Evolution of Cooperation.” Science 211 (1981): 1390–1396.
Johnson, Paul E. “Simulation Modeling in Political Science.” American Behavioral Scientist 42, no. 10 (1999): 1509–1530.
Page, Scott E. “Computational Models from A to Z.” Complexity 5, no. 1 (1999): 35–41.
Schelling,Thomas C. Micromotives and Macrobehavior. New York:W.W. Norton, 1978.
Taber, Charles S., and Richard J.Timpone. Computational Models. Thousand Oaks, Calif: Sage, 1996.