Using genetic algorithms in dynamic models of speculative attack

The evolution of speculative attack models shows certain progress in developing the idea of the role of expectations in the crisis mechanism. Obstfeld (1996) defined expectations as fully exogenous. Morris and Shin (1998) explain how information matters for obtaining equilibrium in the currency market, however, they did not treat the agents' expectations as fully heterogenous. The dynamic approach proposed by Angeletos, Hellwig and Pavan (2006) operates under a more sophisticated assumption about the learning process that tries to reflect the time-variant and complex nature of information in the currency market much better. But this model ignores many important details like a Central Bank cost function. A genetic algorithm permits avoidance of problems connected with incorporating information and expectations into the agent decision making process to an extent. There are some similarities between evolution in nature and currency market performance. In our paper an assumption about rational agent behavior in the efficient market is criticized. We present our version of the dynamic model of a speculative attack, in which we use a genetic algorithm to define the decision-making process of the currency market agents. The results of our simulation seem to be in line with theory and intuition. An advantage of our model is that it reflects reality in quite a complex way, i.e. the level of noise changes over time (decreasing), there are different states of the fundamentals (with a “more sensitive” upper part of the scale), the number of inflowing agents can be low or high (due to different globalization phases, different capital flow phases, and different uncertainty levels).

JEL Classification: C6 Mathematical Methods; Programming Models; Mathematical and Simulation Modeling, F3 International Finance, and E5 International Relations, National Security, and International Political Economy

Keywords: currency crisis, dynamic model, genetic algorithms