Giorgio ARICI

APPLICATIONS OF INFORMATION THEORY TECHNIQUES IN FINANCIAL FIELD

Prediction problems have been of central importance in communication systems context for information theory, thus, since ‘50s, many studies have been carryed out to employ such tecniques in financial context in order to predict market behavior. However most of these theories assumed the complete knowledge of the process that generated stocks’ prices, which is unrealistic and of scarse utility in trading activity.
The goal of my research project is to review financial metrics of risk (such as Value-at-Risk and Expected Shortfall) in presence of relaxed stastical assumptions and propose alternative methods for choosing an optimal portfolio.

 
ESTIMATION OF THE UNSEEN AND DISTRIBUTION SUPPORT EVALUATION

There exist methods that, given a sequence of length N, return an estimate of the number of unseen symbols (not in the observed sequence) one would see if he observed a new sequence of length m*N.
These techniques return evaluations of the number of symbols over which a distribution is defined (support) and they can also give an estimate of the distributions that is superior than the trivial empirial frequency estimation.


Curriculum: Telecommunication Engineering

Tutor: Marco DALAIRiccardo LEONARDI

email: g.arici005@unibs.it

 

Link to the research group web page

Link to publication list