Paper Info
Title
Multifractal analysis of agent-based financial markets.
Abstract
To analyze financial time series exhibiting volatility clustering, long-range dependence, or heavy-tailed marginals, we exploit multifractal analysis and agent-based simulation. We develop a robust, automated software tool for extracting the multifractal spectrum of a time series based on multifractal detrended fluctuation analysis (MF-DFA). The software is tested on simulated data with closed-form monofractal and multifractal spectra to ensure the quality of our implementation. We perform an in-depth analysis of General Electric's stock price using traditional time series techniques, and contrast the results with those obtained using MF-DFA. We also present a zero-intelligence agent-based financial market model and analyze its output using MF-DFA. We study the changes in the macrolevel time series output as analyzed by MF-DFA when altering one of the microlevel agent behaviors. Finally we explore the potential for validating agent-based models against empirical time series using MF-DFA.
Year
DOI
Venue
2013
10.5555/2675983.2676158
WSC '13: Winter Simulation Conference Washington D.C. December, 2013
Keywords
Field
DocType
software agents,time series
Data mining,Stock price,Computer science,Software agent,Exploit,Software,Detrended fluctuation analysis,Volatility clustering,Financial market,Multifractal system
Conference
ISSN
ISBN
Citations 
0891-7736
978-1-4799-2077-8
2
PageRank 
References 
Authors
0.51
1
2
Name
Order
Citations
PageRank
James R Thompson152.80
James R. Wilson2840143.42