Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.663856
Title: Robustness of irrationality in financial markets
Author: Williamson, Paul
Awarding Body: University of Edinburgh
Current Institution: University of Edinburgh
Date of Award: 1999
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Abstract:
Recent research in financial economics has suggested that models incorporating agents whose expectations are not fully rational may help us to understand and explain financial market behaviour. Using a noise trader framework this thesis presents a theoretical appraisal of the robustness of such irrational beliefs and the prescriptions of noise trader models. In contrast to previous work we show that the effects of noise trading are not a short-term phenomena. As agent's horizons increase, noise trader induced risk becomes more important particularly if their beliefs are persistent. Various proposed policy measures to reduce speculative behaviour are considered. We demonstrate that noise trading is robust to the imposition of both linear and non-linear transaction taxes, but could be reduced by direct market intervention. A new mixed policy of tax financed intervention is shown to reduce the destabilising effects of noise traders. We also consider models in which all agents are endowed with adaptive learning rules. Conditions for convergence of the adaptive model to the rational expectations solution and expectational stability are derived. An evolutionary version of the model is investigated through computer experiments. Asset prices converge to an neighbourhood of the rational expectations equilibrium but exhibit exaggerated movements consistent with empirical evidence. The experimental evidence accords with the analytical results on the robustness of noise trader beliefs and their effects, but suggests that noise traders will not come to dominate the market.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.663856  DOI: Not available
Share: