Comparative analysis between standard risk measurements and behavioral simulations
Tutor/a - Director/a
Estudiant
Urbano García, Francisco
Tipus de document
Projecte Final de Màster Oficial
Data
2012
rights
Accés obert
Editorial
Universitat Politècnica de Catalunya
UPCommons
Resum
This project explores behavioral driven simulations as an alternative to the existing classical
methods to calculate the most common risk measurements for financial time series, that is, VaR
(value at risk) and Expected Shortfalls.
. This Final Master Research Project focuses in a comparative analysis among GARCH family models forecast power and a behavioral based simulation of traders in the stock market. The simulation target is to explain and give insights on why the volatility in financial series behaves the way it does, thus, this simulation goes beyond a mere formal model fitting of the data and tries to explain the mechanics within the data
