Efficient Simulation of Financial Stress Testing Scenarios with Suppes-Bayes Causal Networks

Gelin Gao, Bhubaneswar Mishra, Daniele Ramazzotti

Research output: Contribution to journalArticle

Abstract

The most recent financial upheavals have cast doubt on the adequacy of some of the conventional quantitative risk management strategies, such as VaR (Value at Risk), in many common situations. Consequently, there has been an increasing need for verisimilar financial stress testings, namely simulating and analyzing financial portfolios in extreme, albeit rare scenarios. Unlike conventional risk management which exploits statistical correlations among financial instruments, here we focus our analysis on the notion of probabilistic causation, which is embodied by Suppes-Bayes Causal Networks (SBCNs), SBCNs are probabilistic graphical models that have many attractive features in terms of more accurate causal analysis for generating financial stress scenarios. In this paper, we present a novel approach for conducting stress testing of financial portfolios based on SBCNs in combination with classical machine learning classification tools. The resulting method is shown to be capable of correctly discovering the causal relationships among financial factors that affect the portfolios and thus, simulating stress testing scenarios with a higher accuracy and lower computational complexity than conventional Monte Carlo Simulations.

Original languageEnglish (US)
Pages (from-to)272-284
Number of pages13
JournalProcedia Computer Science
Volume108
DOIs
StatePublished - 2017

Fingerprint

Risk management
Testing
Learning systems
Computational complexity
Monte Carlo simulation

Keywords

  • Causality
  • Classification
  • Decision Trees
  • Graphical Models
  • stress Testing
  • Suppes-Bayes Causal Networks

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Efficient Simulation of Financial Stress Testing Scenarios with Suppes-Bayes Causal Networks. / Gao, Gelin; Mishra, Bhubaneswar; Ramazzotti, Daniele.

In: Procedia Computer Science, Vol. 108, 2017, p. 272-284.

Research output: Contribution to journalArticle

@article{b9b8a8a423f9410eaa1c4b8313b0cdb3,
title = "Efficient Simulation of Financial Stress Testing Scenarios with Suppes-Bayes Causal Networks",
abstract = "The most recent financial upheavals have cast doubt on the adequacy of some of the conventional quantitative risk management strategies, such as VaR (Value at Risk), in many common situations. Consequently, there has been an increasing need for verisimilar financial stress testings, namely simulating and analyzing financial portfolios in extreme, albeit rare scenarios. Unlike conventional risk management which exploits statistical correlations among financial instruments, here we focus our analysis on the notion of probabilistic causation, which is embodied by Suppes-Bayes Causal Networks (SBCNs), SBCNs are probabilistic graphical models that have many attractive features in terms of more accurate causal analysis for generating financial stress scenarios. In this paper, we present a novel approach for conducting stress testing of financial portfolios based on SBCNs in combination with classical machine learning classification tools. The resulting method is shown to be capable of correctly discovering the causal relationships among financial factors that affect the portfolios and thus, simulating stress testing scenarios with a higher accuracy and lower computational complexity than conventional Monte Carlo Simulations.",
keywords = "Causality, Classification, Decision Trees, Graphical Models, stress Testing, Suppes-Bayes Causal Networks",
author = "Gelin Gao and Bhubaneswar Mishra and Daniele Ramazzotti",
year = "2017",
doi = "10.1016/j.procs.2017.05.167",
language = "English (US)",
volume = "108",
pages = "272--284",
journal = "Procedia Computer Science",
issn = "1877-0509",
publisher = "Elsevier BV",

}

TY - JOUR

T1 - Efficient Simulation of Financial Stress Testing Scenarios with Suppes-Bayes Causal Networks

AU - Gao, Gelin

AU - Mishra, Bhubaneswar

AU - Ramazzotti, Daniele

PY - 2017

Y1 - 2017

N2 - The most recent financial upheavals have cast doubt on the adequacy of some of the conventional quantitative risk management strategies, such as VaR (Value at Risk), in many common situations. Consequently, there has been an increasing need for verisimilar financial stress testings, namely simulating and analyzing financial portfolios in extreme, albeit rare scenarios. Unlike conventional risk management which exploits statistical correlations among financial instruments, here we focus our analysis on the notion of probabilistic causation, which is embodied by Suppes-Bayes Causal Networks (SBCNs), SBCNs are probabilistic graphical models that have many attractive features in terms of more accurate causal analysis for generating financial stress scenarios. In this paper, we present a novel approach for conducting stress testing of financial portfolios based on SBCNs in combination with classical machine learning classification tools. The resulting method is shown to be capable of correctly discovering the causal relationships among financial factors that affect the portfolios and thus, simulating stress testing scenarios with a higher accuracy and lower computational complexity than conventional Monte Carlo Simulations.

AB - The most recent financial upheavals have cast doubt on the adequacy of some of the conventional quantitative risk management strategies, such as VaR (Value at Risk), in many common situations. Consequently, there has been an increasing need for verisimilar financial stress testings, namely simulating and analyzing financial portfolios in extreme, albeit rare scenarios. Unlike conventional risk management which exploits statistical correlations among financial instruments, here we focus our analysis on the notion of probabilistic causation, which is embodied by Suppes-Bayes Causal Networks (SBCNs), SBCNs are probabilistic graphical models that have many attractive features in terms of more accurate causal analysis for generating financial stress scenarios. In this paper, we present a novel approach for conducting stress testing of financial portfolios based on SBCNs in combination with classical machine learning classification tools. The resulting method is shown to be capable of correctly discovering the causal relationships among financial factors that affect the portfolios and thus, simulating stress testing scenarios with a higher accuracy and lower computational complexity than conventional Monte Carlo Simulations.

KW - Causality

KW - Classification

KW - Decision Trees

KW - Graphical Models

KW - stress Testing

KW - Suppes-Bayes Causal Networks

UR - http://www.scopus.com/inward/record.url?scp=85027328213&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85027328213&partnerID=8YFLogxK

U2 - 10.1016/j.procs.2017.05.167

DO - 10.1016/j.procs.2017.05.167

M3 - Article

VL - 108

SP - 272

EP - 284

JO - Procedia Computer Science

JF - Procedia Computer Science

SN - 1877-0509

ER -