Modeling fund and portfolio risk: A bi‐modal approach to analyzing risk in turbulent markets

• Author: D. Sykes Wilford,Iordanis Karagiannidis
• DOI: http://doi.org/10.1016/j.rfe.2015.02.005
• Date: 01 April 2015
• Published date: 01 April 2015

Modeling fund and portfolio risk: A bi-modal approach to analyzing risk

in turbulent markets☆

Iordanis Karagiannidis

a

,D.SykesWilford

b,

⁎

a

BastinFinancial Leadership Lab,The Citadel, United States

b

Businessand Finance, The Citadel,United States

abstractarticle info

Availableonline 7 March 2015

JEL classification:

G11

G12

G13

G17

Keywords:

Risk analysis

Monte Carlo

Portfoliodiversification

Commodity futures

Following the financial crisisof 2008, it has been argued that Value at Risk (VaR), andrisk analysis in general,

failed to alert risk managers of the turbulence on the horizon. This is a misguided view that should not have

come as a surprise because many widely circulated academic papers and discussions suggested, well before

the crisis, that simpleVaR results could easily be misinterpreted if the circumstancesfor its proper use are not

fully understood. This paper address es some ways in which VaR concepts may be applied more effectively.

Non-standard Monte Carlosimulations are utilized. Whereas standard mean–variance definedmethodologies

usingMonte Carlo analysismay not capturehow “fat”a lower tailmay actually be, a bi-modalswitching structure

betweenassumed normalperiods and possibleturbulent economicperiods may help resolvethe problem. Lower

boundaries (worst case paths) of the different (normal versus bi-modal) pro cesses are mapped to illustrate

implied riskiness of portfo lios if turbulence occurs. The analysis implies th at no mechanical risk analysis is

sufficiently divorced from a judgment call about possible market disruptions; however, a bi-modal approach

allows quantification of the saidjudgment in conjunctionwith empirical observationsfrom history.

© 2015 ElsevierInc. All rights reserved.

Following the financial crisis of 2008 and 2009 it has been argued

that Value at Risk (VaR) , and risk analysis in general, have fai led the

profession and the ec onomy by misleading profe ssionals to underes-

timate the risks in the system. Its usefulness is in doubt and major

“fixes”have been suggested. Interestingly, arguments against VaR

are not new. Many widely circulated papers and discussions argued

well before the crisis that simple VaR could be very misleading.

1

As

such the degree to which rating agencies and regulators were wed-

ded to a simplistic approach to risk analysis is even more troubling.

Claims became commonplace that Markowitz approaches to risk

diversification and supposed failures of VaR like methodologies

created the crisis .

With respect to Markowitz (1952, 1959) thes e claims seem total-

ly unfair. When a Dean of a m ajor Business School deno unced Marko-

witz contribution suggesting that he give back his Nobel prize in

econom ics, it underscored how poorly thes e basic theories that

Markowitzintroduced were understood.The misperceptions led to de-

pendence on simplistic risk modeling,resulting in simple-minded ap-

proaches that would no doubt lead to falseconclusions. It was not the

faultof Markowitz theory norwas it for a lack of academicmaterial sug-

gesting how that theory and risk analysis should be utilized correctly

that triggeredthe crisis.

2

The culprits mos t accused of “causing”thecrisis by the media were

the mortgage back securities (MBS) industry,hedge funds, structured

investment vehicles (SIVs), and of course the banks. At the h eart of

the crisis was home mortgages. So what role then did VaR play, even

in simple form?

Followingthe LTCM (Long Term CapitalManagement)crisis of 1998,

many managers (in particular those familiar with both MBS manage-

ment and Markowitz theories) argued th at a simplistic VaR, based

upon a normaldistribution, would lead to falseestimates of actual risk

of MBS funds, certain debt portfolios, deb t-inked hedge funds and

othersimilar financialstructures. The logicwas simple. Debtis effective-

ly an option on the value of the firm(or houses) and as such entitiesthat

buy the debt arewriting said options. Historybased return estimations

utilized to calculate the risk associated with any structurethat is debt-

Reviewof Financial Economics 25 (2015)19–26

☆Wilfordwishes to thank The CME Foundationfor the supportof some of the underly-

ingresearch. The authorsthank BlufordH. Putnam for his input. Anyerrors are those of the

authors.

⁎Correspondingauthor.

E-mailaddress: wsykes@laudisi.com(D. Sykes Wilford).

1

See, among others, Alexander and Sheedy (2008),Pritsker (2006),Putnam et al.

(2002),Taleb (2008), and Angelidis and Degiannakis (2007). These references discuss

shortcomingand/or provide good backgroundto show that the issues of mis-application

of VaR approacheswere known well beforethe crisis in the literature.

2

Wilford (2012)discusses the issue of misinterpretationof Markowitz portfolio opti-

mization at length as well as some o f the dangers of a simplistic application of the

basic tenants to risk analytics. Many of these issues are discussed in various sectionsof

Markowitz(1987).

http://dx.doi.org/10.1016/j.rfe.2015.02.005

1058-3300/©2015 Elsevier Inc. All rightsreserved.

Contents listsavailable at ScienceDirect

Review of Financial Economics

journal homepage: www.elsevier.com/locate/rfe