401-412

• Dima, Bogdan
• Pirtea, Marilen Gabriel
• Cristea, Ştefana

A critical issue in financial markets’ research is the debate between the academic orthodoxapproach of the Efficient Markets’ Hypothesis and the critics rising from the behavioral finance paradigm andpractice. Several alternative explanations have been proposed in order to provide a more realistic description ofthe financial markets’ inner mechanisms. This debate’s importance consists in the implications of the adoptedpoint of view on the assessment of the financial markets’ predictability degree. [4], [5] proposed a unifiedapproach labeled as Adaptive Markets Hypothesis. Thus, from a practical point of view, the main issue consistsin providing a pertinent answer to the next question: Is an active portfolio management able to provide betterresults that a passive “follow the market on long-run” strategy? If in adaptive models the markets are consideredto display, at least in a certain sense, a degree of predictability, then, one of the major difficulties in supplyingempirical evidences is the requirement of knowing ex ante the “exact” forecast model used by the economicsubjects. The aim of this study is providing a solution to this problem inspired by the transduction’s (supervisedlearning) algorithms. Our main output consists in the thesis that the forecasting errors matter for price formationin financial markets. So, despite the fact that nor the theoretical foundations nor the empirical evidences areconclusive, we argue that the nature of the “exact” learning mechanisms can be seen as one of the key variablesin investors’ decisions and markets evolution. There is a significant positive payoff of a more detailed study ofsuch mechanisms inside an extended framework of financial markets as complex systems

• Economics and Business

• Οικονομία
• Financial management
• Finance
• Business management

financial markets, FTSE 100, adaptive markets hypothesis, forecasting algorithms, forecasting errors, adaptive mechanisms

Περιέχει γραφήματα, πίνακες και βιβλιογραφία