Year: 2003
Type: dissertation
Abstract: In almost all cases a decision maker cannot \nidentify ex ante the true process. This observation has led \nresearchers to introduce several sources of uncertainty in \nforecasting exercises. In this context, the research reported in \nthese pages finds an increase of forecasting power of financial time \nseries when parameter uncertainty, model uncertainty and optimal \ndecision making are included. The research contained herein evidences \nthat although the implementation of these techniques is not often \nstraightforward and it depends on the exercise studied, the \npredictive gains are statistically and economically \nsignificant over different applications, such as stock, bond and \nelectricity \nmarkets.
Author: Francesco Ravazzolo
Language: English
Field: Decision Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W1570009503