Search for content and authors
 

Heterogeneous Agent Models With Threshold-Induced Switching

Harbir Lamba 

George Mason University (GMU), Fairfax 22030, United States

Abstract
We consider a class of financial market models that provides a consistent framework within which the effects of different types of agent motivations and behaviours can be systematically studied. The agents operate within a highly simplified environment where they are only able to be long or short one unit of an asset. The price of the asset is influenced by both an external Brownian information stream and the aggregate demand of the agents. The (evolving) strategy of each agent is represented by a pair of moving thresholds straddling the current price. When the price crosses either of the thresholds for a particular agent, that agent switches position and a new pair of thresholds is generated.

Differing rules governing threshold evolution can mimic different sources of investor motivation, running the gamut from purely rational information-processing, through rational (but often undesirable) behaviour induced by perverse incentives and moral hazards, to purely psychological effects. The simplest forms of threshold dynamics coincide, both practically and philosophically, with efficient market models operating under the rational expectations assumption --- agents trade due to their differing future expectations but, since no coupling occurs between agents, the efficient market price is always achieved. However, in the general case, coupling between agents can occur causing significant asset mispricing.

The inclusion of a herding propensity is especially interesting since it is suspected to be an important factor in many past financial crises. Within this class of models, herding induces bubbles and crashes and the price-return data display power-law behaviour in the tails with exponents closely matching those observed in real markets. Furthermore, since the herding behaviour can be introduced into an otherwise efficient-pricing market, a direct causal relationship between herding and fat-tails can be established. We shall conclude by considering possible causes of volatility clustering and outlining a mathematical approach for quantifying such deviations from the traditional rational expectations assumption.



 

Legal notice
  • Legal notice:
 

Presentation: Oral at International Conference on Economic Science with Heterogeneous Interacting Agents 2008, by Harbir Lamba
See On-line Journal of International Conference on Economic Science with Heterogeneous Interacting Agents 2008

Submitted: 2008-03-11 05:03
Revised:   2009-06-07 00:48