Is the Efficient Frontier Efficient?

by

William C. ScheeL William J. Blatcher, Gerald S. Kirschner, John J. Denman

Abstract
The paperdefinesplausibleways to measuresamplingerror within efficient frontiers, particularly when they are derivedusing

dynamicfinancial analysis. The propertiesof an efficient surfaceare measuredboth usinghistorical segmentsof dataand

using bootstrapsamples. The surfacewas found to be diverse,and the compositionof assetportfolios for points on the

efficient surfacewas highly variable.

The paperbacesperfOrn1ance
of on-frontier and off-frontier investmentportfolios for different historical periods. Therewas

no clearcut superiorityto the on-frontier setof portfolios, althoughlower risk-return on-frontier portfolios were generally

found to perform betterrelative to comparable,off-frontier portfolios than thoseat higher risk levels. It is questionable

. whetherpracticaldeploymentof optimization methodscanoccur in the presenceof both high samplingerror andthe relatively
inconsistenthistorical performanceof on-frontier portfolios.

The implications of this paperfor DFA usageof efficient frontiers is that samplingerror may degradethe ability to effectively

distinguishoptimal and non-optimalpoints in risk-return~. The analystshouldbe cautiousregardingthe likelihood that

points on an efficient frontier are operationallysuperiorchoiceswithin that space. Thereare many possiblefrontiers that

optimally fit different empirical samples. Samplingerror amongthem could causethe frontiers to traversedifferent regions

within risk-rebun space,ped)apsat points that are disparatein a decisionsense. What is an efficient point on onefrontier may

1 William C. Scheel,Ph.D., Consultant,DFA Technologies,LLC; William J. Blatcher,Financial Actuary, AEGIS Insurance

Services,Inc.; GeraldS. Kirschner,FCAS, MAM Vice President,ClassicSolutionsRisk Management,Inc.; and JohnJ.
Denman,SeniorVice President,Finance,AEGIS InsuranceServices,Inc. The authorsgratefully acknowledgethe very
constructivecommentsof reviewersof the paper.

PageI of 45
be inefficient when calculated from a different sample. The paper finds the use of an efficient surface to be helpful in

diagnosing the effects of such sampling error.

1.0 Introduction

Companieschooseamonginvestmentsoften with the purposeof optimizing somegoal and alwayslimited by constraints.

Assetsare divided amongcompetinginvestmentalternativeswith the hopethat risk will be minimized for a desiredlevd of

return,either investmentreturn or overall return. Whenthe allocationfulfills the goalswithin the boundariesof constJaints,it

is thoughtto be efficient. The allocation is deemedto be a ~mber of the efficient setat a point on an efficient frontier. It is

efficient becauseit dominatesoff-frontier, interior points in the risk-return space.

This paperinvestigatesthis popular investmentallocation strategyin two ways. First. it seeksto detenninewhat the sensitivity

of the frontier is to possiblesamplingerror in risk-returnS18:e. Secondly,both on-frontier and off-frontier portfolio

allocationsfor actual seriesof returnsare b'ackedfor their respectiveperfonnance. We begin with an apologue;it gives the

reader both a rationale and definition of what we mean by an efficient surface.

1.1 A Sampling Error Apologue

I walk into a casinowith shakykneesand a rathersmall~. Betting doesn'tcomeeasilyfor me, and I expectto losethe

stake. Ralphtold me I would lose it But. I havea bevy of infonnation gleanedfrom experimentsRalph did with a

computerizedsimuJationof a Clapstable. One of the items I call "knowledge" is the efficient surfacehe madefor me. Ralph

said it would help me understandthe risk/rebJrnpropertiesof the crapstableand guide me in allocating my stakeamongthe

variousbetsthat I can make

"There are manybets you can makeat the table," Ralph explained. "'Come', 'Big-8' am lots of others. think of the gaming

as a multivariate~. Of course,it hasprobabilitiesthat are objectiveand canbe measured Do you want me to figure out

the combinatoricsof the crapsgameaM derive analytic solutionsfor optimal bet placement?My consulting fee might be a bit
high becausethe math will take awhile, but I could do il ..

mentallyrecalculatedmy meagerstakeand replied, "Is therea lessexpensivewayT'

Page2 of 4S
Ralph shruggedand said,"Sure. I can usea computersimulationI haveand take a sampleof gameoutcomes. I'U usethe

sampleto empirically developa covariancematrix for someof the bets. Then. I'll figure out which combinationsof bets have

minimum variancefor a particular payoff. You can choosewhich risk/return profile of bets is bestfor you. You'O be able to

more efficiently allocateyour stake. By-the-way, this is called an efficient frontier-it gives a profile of bets that areexpected

to producea given return with minimum variance. I'll do a sampleof 25 gameseachwith a combinationof variousbets. This

will keep the cost down.

"Well, okay," I replied. "But, will this singleefficient frontier really work?"

"What do you mean,'single frontier'?" he asked.

"What if the sample your computer simulation comes up with is unusual?" Ralph scratched his head, and I continued, "You

measure this thing you call a sample covariance matrix. But, what if you took a different sample? You'd get a different

sample covariance matrix, right?"

"Yes."

" And, it might be different?"

"Yes. Even materially different."

"So, your efficient frontier (Ef) is subjectto samplingelTor-it was empirically derivedfrom the sampleof only 25 games." I

then asked, "What if you had a second sampleof 25 games and did another mathematical optimization. So, we now have two

different EFs; both do the samething, but the answersare different Which one do I usewhen I walk into the casino?"

Ralph exclaimed,"I'll take a sample,and then another,and another. Eachwill havea different EF. Then, I'll plot eachpoint

of the samples'EFs in risk/return ~ce. I'll count the numberof times the variousEFs traversea particular cell in that space.

Page3 of 45
Maybe 10 EFs b'aversethe cell at the coordinates(10,15). Maybe only 3 EFs traversedthe cell at (1,3). Don't you see? Just

by countingthe numberof times the sampleEFs baVersea region in risk/retmn spaceaOOnormaliring the count to

probabilities, I can measure an efficient surface.

I asked,"Why is the surfaceimportantr'

Ralph was now animated. He leapedto his feet. "Because,if the varioussampleEFs all nversed the samecells, the EFs

would all be the ~e-d1ere would no samplingerror. What if the surfaceis spreadout? SUA>Ose
somesectorsof it are

relatively flat? Then the efficiency of the EFs varies. Would you prefer to pick a point on the surface(with a particular

combinationof bets)that appearsmost often amongdifferent EFs? Probablyyou would. You want the surfaceto be tightly

peaked In 3-dimensions,that'sa ridge or very JXJintyhill; in two dimensions,it is a probability distribution with little

variance.

He then went home to begin the chore of sampling and constructing an efficient surface for me. I began to think, ..A single

efficient frontier is measuredfrom data. We often think of the databeinga samplefrom a replicableexperiment. If a sample

of dice gamesis observed,the n-tuple bet outcomesfor the correlatedbetsis the empirical data sourcefor an optirni7J11ioo.
It is

easyto seehow different samplescan be drawn when talking aboutdice games. But, the world of security returnsis different

from a crapstable. What is a samplethere? What is the meaningof samplingerror, and how might it affect the way I measure

efficient frontiers? Would an EF for securitiesreally be efficientr

Theseare imponant questions-ones addressedin this paper. It is difficult to think of how we'd repeatan experiment

involving securityreturns. Is a seriesof experimentsonethat usesdifferent historical periodsof returns? Is it a bootStI3pof a

broad segmentof history? Theseare the two approachesthat are equivalentto samplingand measuringsamplingerror. The

resultof our measurements

is an efficient surface.

1.2 Roadmap for the Paper

Section2 of the ~r lays the groundworkfor measuringsamplingerror that affectsefficient frontier measurementWe

examinetwo approachesthat seem~cularly useful for dynamicfmaocial analysis(OFA). We also review the literature

P.4of4S
relatingto EF efficiency. Section3 introducesdie notion of an efficient su1f«e-this is a constructfOf understaOOing
and

measuringsamplingerror in EFs. In this sectionwe describethe methodologyand data setusedin our ggdy

The main body of resultsis presentedin Sections4, 5 and 6. We measureforecastperfonnanceof efficient frontiers in Section

4. We areparticularly concernedaboutthe performanceof off-frontier portfolios. Are they really inefficient? Do on-frontier

portfolios dominateperformanceas we might antici~te given that they are billed as "efficient'" The evidencewe presentin

Section4 showsinstability in EFs derivedboth with historical segmentsand [Link] leadsus to conclucklater

that cautionshouldbe exercisedwhen using efficient frontiers in OFA analysis.

On the road to this conclusion,we closely examinethe efficient surfacein Section5. It portrayssamplingerror from two

different ~ve&-historical and bootstrap sampling. The efficient surface is a useful COostJ\lctfor visualizing sampling

error in EFs. We observethat such~r is particularly large in the high risk/returnregionsof the surface This OOservationis

reinforcedin Section6 by observingthe diversity of portfolio compositionas we comparedifferent historical segments.

The final sectionis devotedto conclusionsand cautionson the ~ of EFs in OFA work. We concludethat EFs may not

warrant the terD1efficient. Their best use may be as advisory measurements concerning the properties of risk/return space.

2.0 Scenario Generationin DFA

Dynamic financial analysisinvolves [Link] manytypesof scenariosthat are simulatedso that the

modelbuilder can measurea hypotheticalstate-of-the-worldwith accountingmetrlcs. Assetgeneratorstypically createreturns

for [Link] model exogenouseconomicconditions. Eachmodelerseesthe forcesof the financial markets

unfolding accordingto a setof roles. The rule setis almostasdiverseas the numberof modelers.

SomeOFA modelbuilders prefer stochasticdifferential equationswith variousdegreesof functional [Link]

transitionof returnsover time, aswen as the correlationsamongdifferent assetcomponents,alwaysis representedin multiple

[Link] DFA modelersusemultivariatenomlai [Link] conjecturea covariancematrix of

investmentreturns. Thesemodelsdo not havetime-dependentb'ansitiODmodelinginformation. Suchan efficient frontier, by

Page5 of 45
definition. hasno time transition properties. A sampletakenfrom any sub-periodwithin the time-serieswould contain

samplingerror, but otherwise,the investmentallocation would be unaffected

Both approachesbegin with a single instanceof reality. They both pIrport to model it One [Link],

useslargely subjectivemethodsto ~eterize the process.2Another awroacb to modelingclings to assumptionsthat seemto

be or are takento be realistic! Both producescenariosthat are deemedsufficiently similar to reality to representit for the

pwposeat hand

The efficient frontier calculationcanbe a constrainedoptimization either basedon a SUDpiefrom a historic seriesof returnsor

a derivedserieswith smoothingor other ad hoc lkijustment Alternatively. an EF may be createdfrom simuJatedOFA results.

Both the efficient frontier aM OF A asset-basedmodeling are using the same set of beliefs regarding the manner by which

statisticallyac<:eptable
parametersare used.4They both stan with a singlehistoric time-seriesof rehJrDS
for variouscomponent

assets.

2.1 Two Viewpoints on the Use of Efficient Frontiers

The pmctitionerhasa straightforwardobjective: define investmentallocationstrdtegygoing forward. Today's portfolio

allocationleadsto tomorrow's result The portfolio is then rebalancedrelative to [Link] new one leOOs
to new

results. The cycle repeats. Where doesthe chickenend and the egg begin? In practice,the practitionerhasonly one instance

of yesterday'sreality and tomorrow's expectationsfrom which to constructa portfolio and a model.

There are at least two approaches to using a OF A model to define an investment allocation. In one, a OF A analyst might set up

an initial allocationof assetsusing an efficient frontier obtainedfrom quadIaticoptimi7Jitionon a prior historical period. A

OFA model would be repeatedlynm-a different state-of-the-worldwould ensueeachtime, and a different ~g obtained

2The calibration may dependon examinationof stylistic facts,but thereseldomis formalized. statisticalhypothesistestingto
judge whetherthe facts canbe acceptedas suchor whetherthe representationof thesefacts in the model is really a scientific
dctermination.
.1 Somemodelsusemultivariate nonnal simulationfor renderinginvestmentreturnsfor consecutiveperiods. Thereusua1Iy is
an assum~on that the covariancematrix usedfor multivariate nomlal simulationis stationaryfrom period to period in these
models.
4 DFA and optimization do havea critical junction. SomeDFA modelersbelievethey understandtime dependencies within
period-to-periodratesof return. EF attem~ to optimize expectedreturn. If there is a time-<iependence
[Link] should
be factoredinto the expectedreturnsusedto build the EF for any period.

Page6 of 4S
for the metric. Thesesimulationsproduceendpointsin the modeledrisk-return space. In this approach,one beginningasset

allocation leads to many different observations about eoopoints. The reason they are different is that, although each starts with

the samestate,the model simulatesvarious [Link] probablyleadsto a different endpointfor the

planninghorizon.

But, anotherviewpoint exists.5 We refer to it asthe hybrid approach. Supposethat history servesa valid purposein calibrating

a model,but shouldnot be usedto define a beginningallocation. In this viewpoint, the investmentmix is suggestedby the

optimizer. DFA servesonly to measurewhat could happenwith somehypotheticalstartingallocation.

The optimizer dealsthe cardsin this deck, and DFA traceswhere the cardslead.6.7The optimizer, not the modeler,submitsan

initial allocationfor review. In this hybrid [Link] is no initial portfolio basedon optimization using prior history. In

the hybrid model,the optimizer finds a portfolio, which leadsto an expost optimal result The metric usedin this optimization

is part of the DFA model-it is calculatedby the accountingmethodologyof the model as it generatesfuture statesof the

world. It may be difficult to reconcilethe useof efficient frontiers for investmentswithin hybrid-DFA modeling that, on the

one hand,believesthere is a historically dependentcomponentthat canbe usedfor calibration,but rejectsthe useof datato

define a startingportfolio. Yet. on the other hand,simulationsof that model are derivedto constructan efficient frontier. It

may appearas thoughhistory hasbeenrejectedas information for the purposesof decision-making,yet indirectly it is usedto

representthe future. The startingportfolio in the hybrid approachis basedat leastindirectly through modelingand should

representan analyst'[Link] in theory built into the model for return scenariogenerationand

that modelwas calibratedto history in somefashion.

In DFA work. a performancemetric is chosen. This metric is measuredwithin a risk-return space. The metric mustbe

measurableaccordingto the chosenaccountingframework. Risk might be variance,semi-varianceor somecbance-constrained

5Correnti,et ai, review an approachsimilar to the hybrid model describedhere.

6The optimizerpositsa trial solution; it consistsof a certainportfolio allocation. This trial allocation doesnot dependon any
prior allocationof [Link] ensuesduring the optimizationperiod (and underthe control of the OFA model)
also is unknownto the optimizer. The objectivevalue that is returnedby the model is driven by the initial triaI solutionand
model machinationsthat build on the trial solution.
7 Investmentratesare forecastedby the OFA model, which might use multivariate normal simulation. There may be an

overlapbetweenwhat the optimizer usesand what the OFA model uses. For example,the covariancematrix usedfor the
multivariatenormal simulation is estimatedfrom historical dataand generallyis assumedto be stationaryduring the forecast
period. It is usedboth by the optimizer and by the OFA model.

Page7 of 45
function of the metric. In the real world. the corporate manager is rewarded for favorable performance of the metric and often

penalized by unwanted risk in the metric. The volume of investment in various stochastic components affects a metric's

performance. The operational question is how should an allocation be made in investments so that performance of the metric is

optimized.

In the forecastperiod, the modelergeneratesa scenarioof unfolding ratesof return using, say,a multivariate,time-dependent

assetmodel. An examplewould be any of the multi-factor meanreversionmodelsin usetoday. The simulatedprogressionof

returnsfor a scenariogeneratedby one of thesemodelsis affectedby an underlying mechanismthat forcesunusualdeviations

in the pathback towardsan expectedtrajectory of returns. The OFA model typically ties in someway the businessoperations

to the [Link] economicscenariotypically generatesother economicratessuchas the rateof

inflation. A scenariothat is generatedby the economicmodel is takento be exogenous;it is mingled with expectationsabout

corporateperfonnance. The company'soperationsare tied to the exogenousinfluencesof the economicscenario.

In the end,this modeling processis repeatedmany times for the optimizer in the hybrid model. The optimizer requiresan

answerto the question:given an initial [Link] is the end-horizonperfonnanceof the metric. The

optimizer forcesthe model to measurethe result of a simulation experimentgiven only an initial investmentallocation. The

model takesthe allocation and producesan experimentalpoint in risk return space. All that is requiredof the model is its

ability to measurethe trajectoryof the metric within the company'sbusinessplan anda beginningallocation of [Link] this

regard,the hybrid model is using a sort of dynamic programmingapproachto optimization. The possibleoutcomesare

considered,andthe most desirabletracedback to the inputs (initial allocation). The hopeis that the optimizedfeasiblesetis

robustrelative to possiblestochasticoutcomesin the model trajectory. The efficient frontier tracesthe allocationsnecessaryto

achievevarious points in this risk-return space. All of this raisesthe thorny questionof subsequentperformancedominaIK:eof

the on-frontier portfolios in the hybrid model. Do EF points truly dominatethe perfonnanceof off-frontier frontier points-

portfolios that are thoughtto be inefficient and havehigher risk for the samereturn level?

8 A typical behavioralpatternfor businessgrowth is modelingit as a function of inflation, which was generatedby the
economic scenario. Another is to tie severity in claims to underlying inflation as unfolded in the economic model simulation.

Page8 of 45
The reason that this is a hybrid approach is that DF A modeling is not deployed on an optimal asset allocation derived directly

from the prior time-series. Rather, DF A is combined with optimization to answer the single question: how should the portfolio

be immediately rebalanced to achieve an optimal point in risk-return space over the future OF A planning horizon.

Two portfolios can be devised through optiJIIization procedures-one is based on historical results prior to the start of the

simulated future time periods. Another one involves allocations that are selected and tried by the optimizer-the OF A model is

integral to this second approach. The latter hybrid optimization uses DF A-measured metrics in the optimizer goal function. If

applied over the course of the simulated future time periods, and according to the plan of the DF A model, the hybrid approach

woald seem to yield optimal results at the end of the simulated time horizon. There is no reason to suppose that these two

approaches produce the same initial portfolios. Which one is the real optimwn?

During the planning horizon, the hybrid model may ignore imperfections that. in real life, might have (and, probably would

have) been dealt with by on-going decision making. The EF could have been recalculated with realized data and the portfolio

rebalanced. The published state-of-the art in DFA modeling is unclear in this regard; but, it may be that no inua-period

portfolio optimization is done by OF A models between the time the analysis starts with an allocation posited by the optimizer

and when it ends, say, five-years later with a OF A-derived metric. It is inconceivable that an organization would mechanically

cling to an initial, EF-optimal result for an operational period of this length without retesting the waters.

2.2 Limitations of this Study for Use of the Efficient Frontier in DFA

We do not do a completeDFA analysis-there is neithera liability componentnor a conventionalDFA metric suchas

economicvalue of a businessenterprise. Rather,the dataare limited entirely to marketable,financial [Link],we

believe our findings are of value to OFA work. If the efficient frontier producedsolely within a traditional investment

9Thereis no reasonother than a few computationalprogrammingcomplexitieswhy intra-periodoptimizationscannotbe done

within DFA models. The questionis whetherthey are, or they are not, being done. For example,the DFA model can simulate
a wide variety of rebalancingstrategiesincluding the real-life one that involves a rebalancingtrigger for simulatedportfolios
whoseallocation hasdeviatedfrom a recentEF by someamount Mulvey, etal [1998, p. 160] describean n-period simulation
wherein suchrebalancingis triggered. In addition. Mulvey, et ai, describethe useof optimization constraintsin a cleverway
to achievean integrationof strategic,[Link] short-termtactical objectives. However, a DFA modelthat
allows intra-periodoptimization must also capturethe transactionand tax costsassociatedwith the intra-periodrebalancingand
re-optimization. SeeRowland and Conde[1996] regardingthe influenceof tax policy on optimal portfolios and the
desirability of longer term planning horizons.

Page9 of 45
framework has unstable properties, these instabilities will awly to its use in OF A work were it to be calculated and used in a

similar way.

2.3 Other Investigations of the Efficacy of EF Analysis

Michaud basextensivelyinvestigatedthe useof EFs with particular regardits generalefficacy for forecasting. For example,he

basshown [1998,pp. 115-126]that inclusion of pensionliabilities can substantiallyalter the statisticalcharacteristicsof mean-

varianceoptimization for investmentportfolios.

Michaud's book [1998] examinesefficient frontiers both with respectto their inherentuncertaintyand what might be doneto

improvetheir worthiness. He suggeststhat the effectsof samplingerror may be improvedusing a methodologydescribedasa

resamp/edefficientfrontier. The motivation for somekind of improvementover classicalEFs is that "...optimized portfolios

are 'error maximized' and often have little, if any, reliable investmentvalue. Indeed,an equally weighted portfolio may often

be substantiallycloserto b11eMY optimality than an optimizedportfolio." [Michaud, 1998,p. 3].

The determinationof a resampledefficient frontier is complex;Michaud haspatentedit Although his book exposesthe core

of the methodthat he believesimproveson forecasterror, thereis no empirical evidenceprovided in the book that a resampled

efficient frontier basthis desirableeffect. Interestedreadersare directedto his book. The conceptof an efficient surface

espousedin our paperis built on different constructs. We will readdressthe importantwork of Michaud at a later point in the

paper. We now turn to the definition and measurementof an efficient surface.

3.0 The Efficient Surlace

An efficient frontier consistsof points within risk -return spacethat haveminimum risk for a return. If therewere a time-

stationary,multivariate probability distribution for prior history, then history is a samplefrom it History, therefore,would

[Link] error. 1~

10 If therewere conjecture,the multivariate distribution would be subjective,and the efficient frontier would be the subjective
frontier. A subjectivelyderived EF hasno samplingerror, but it may lose operational appealwhen representedin this manner,
becausesubjectivity requiresdifficult reconciliationwithin a corporate,decision-makingframework.

Page10of 45
The concept of a conditional marginal probability distribution either for return or risk emerges, and it, too, would have

sampling error. We discuss the properties of this marginal distribution. an equi-return slice of the efficient surface, in Section

S.l

Were the instance of reality to be a sample. what is the sampling error?

Figure 1 Comparison of Efficient Frontiers for Different Time Periods

Figure 1 showsefficient frontiers for random5-yearblocks of history. The EFs were derived from monthly returnsbeginning

in January,1988. Each curve in Figure 1 requiresoptimizationsfor a 5-yearhistory of returns. The block of monthly returns

was picked at randomfrom the entire time series. The points along eachEF areobtainedfrom separatepassesthroughthe data

with the optimizer. On eachpass,one of the constraintsdiffers. That constraintis the requirementthat the averageportfolio

returnbe a specifiedvalue in the return domain. The optimizer's objectivefunction is the minimization of varianceassociated

with that portfolio expectedreturn.

EachEF in Figure I consistsof nine points; eachpoint involves a separatequadraticoptimization. For example,one of the

optimizationconstraintsis the portfolio expectedreturn,which is setto an equality condition. Therewere nine different

expectedreturnsusedin the study; one was a monthly return of 0.004. An examinationof the figure at this value showsa point

for eachof the four EFs. An empirically derived covariancematrix was detenninedfor eachof the four time seriesillustrated

in Figure I aswell as for hundredsof othersthat are not shown. Thejuxtaposition of the EFs displaysa tangle of overlapping,

Page 11 of 45
crisscrossingcurves.!! This illustration canbe viewed as samplingwith replacementfrom a historical sample;it is appropriate,

then, to view the figure as illustrative of a probability surface. It is a surfaceshowingthe extent of samplingerror provided

therebasbeena stationary,multivariate distribution of components'retums.12Figure 1 indicatesthat it may be hazardousto

accept any particular segment of history as the "best estimator. This figure showsonly severalof the EF curvesthat build up

an efficient surface. Examplesof efficient surfacesappearlater in Figures8 and 10. The distribution of risk in a cross-

sectionalslice of this efficient surfacealso is reviewedin Section5.

The position and slopeof the EFs in Figure I are wildly different, and were other historical EFs to be included,the complexity

would be greater. This lack of historical stability castsdoubt on the operationalvalidity of a particular efficient portfolio

actuallyproducingoptimal [Link] figure alsohints that off-frontier portfolios may perfonn aswell or betterthan

on-frontier portfolios. We examinethis questionof forecastreliability in detail in Section4.

In additionto the positional changesin EFs over time, there is dramaticchangein portfolio compositionalong the curveof

eachEF in Figure An exampleof the changein portfolio compositionfor EFs appearsin Figures2a and 2b. Each chart is

categorical- a tic maIk on the x-axis is associatedwith oneof nine optimi~tion points. Eachchart showsa stackedarea

renderingof the proportion of an assetcomponentwithin the efficient set. If the readerviews the chart in eitherFigure 2a or

2b from left-to-right, the unfolding change,and possiblecollapse,of a particularcomponentis illustrated. This type of chart is

a useful way to show a component'scontributionto the efficient setmoving along the EF from low-fisk-return to high risk-

return portfolios.

11 Somesegmentsof EFs suchas thoseshownin Figure I canbe [Link] is becausethe quadraticoptimizer could
not identify a feasibleset of investmentalternativesfor all of the averagereturnschosenin the analysis. Thereis a small
probability of overlap of databecausethe 5-yearblocks of returnsusedfor eachEF could haveoverlappingsub-periodsof
time.
12 The populationdistribution is [Link] it is estimatedfrom the historical recordby calculationof an empirical

covariancematrix for eachhistorical block.

Page 12 of 45
 Figures 2a Portfolio Compositions for Different Efficient Frontiers

Efficient Frontier Profile

100%

80%

60%

40%

20%

0%
1 2 3 4 5 6 7 8 9
a=Point

Figures 2b Portfolio Compositions for Different Efficient Frontiers

Efficient Frontier Profile

1~
~
80%
8~1
mLBOOff'
70%
«J% 8HYLD
50% 8lST8
40%
8S&F5
30%
20% 8NTlltD
10% aEAFBJ
0%
1 2 3 4 5 6 7 8 8
e=Point

Thereis faint hopethat the two different EF portfolio compositionsshownin Figures2a and 2b will operationallyprOOuce
the

sameresultwhenput in practice-were this to be a reasonablerepresentationof the effectsof samplingerror, the operational

useof efficient frontiers would be questionable;samplingerror swampsoperationalusefulnessand forecastresponsiveness.

Page13of 45

~
However, another illustration, Figure 3, indicates that if history is a sample from a multivariate distribution. there should be

optimism that the efficient frontier evolves slowly, at least measured in monthly metrics. This figure shows EFs calculated

from consecutive, overlapping historical blocks of time. In this case, the time interval between between consecutive EFs is one

month. The stability deteriorates fastest at higher risk-return levels. The result was found to hold for a wide variety of

consecutive historical blocks starting at various points since 1977. This stability may provide an operational basis for

investing in an on-frontier portfolio and seeing its perfOrmalx:e prevail over off-frontier portfolios. at least for relatively short

planninghorizons.

Figure 3 EFs for Consecutive Time Periods

Efficient Frontier
1

L Risk

Thereare otherways to usethe historical record. The papershortly will turn to the useof the bootstrapas a methodof

measuringsamplingerror. First, the dataand manipulationmethodsare describedin more detail.

3.1 Data Manipulation

This studYusesthe time seriesdescribedin Appendix A: Review of Data [Link] where gapswere presentin the

historical record. the portfolio returns are actual. I 3,14

13 The datarepresentreturnsfor a selectedgroup of [Link] no attemptto filter or smooththe time

series in any way. However, a few gaps in the historical record where interpolated.

Page 14 of 45
The datawere usedin two ways: (1) bootstrapsampleswere madefrom the original time seriesin an attemptto approximate

samplingerror phenomena,and (2) various historical seriesof the datawere usedfor performanceanalysis. The study

examinesperiod segmentation,and the performanceof efficient and inefficient portfolios for different forecastdurations.

3.1.1 Historical Performance Analysis

In this sectionof the paper,datafor an efficient frontier are extractedfor an historical period and usedto evaluatethe efficient
frontier. The on-frontier portfolios are minimwn varianceportfolios found using quadraticprogramming.I 5 Off-frontier

portfolios also were calculated.16The studyis concernedwith whetherthe performanceof off-frontier portfolios really were

inefficient comparedto the performanceof on-frontier portfolios.

3.2 Bootstrap Sampling

The bootstrapsampleof a data set is one with the samenumberof [Link] randomreplacementof every elementby

drnwing with replacementfrom the original set of data. Whenthis processof empirical resamplingis repeatedmanytimes,the

bootstrapsamplescanbe usedto estimateparametersfor functions of the data. The plug-in principle (Efron and Tibshimni,

1993,p. 35] allows evaluationof complexfunctional mappingsfrom examinationof the samefunctional mappingon the

[Link] function 8 =t(F: of the probability distribution F is estimatedby the samefunction of the empirical

"" "
distribution F, 8 = t(F) , where the empirical distribution is built up from bootstrap samples. This technique often is

deployedfor the derivation of errorsof the estimate.

14 One techniquefor deploying efficient frontiers within DFA analysisinvolves removalof actual valuesfrom the dataseries
usedin optimization. Thesepoints in the actualtime seriesmay be deemedabnormalities. The efficient frontier calculation
doesnot useall availabledataor usesthem selectively. SeeKirsclmer [2000] for a discussionof the hazardsof historical
p;riod segmentation.
5 All optimization was doneusing Frontline Systems,Inc. PremiumSolverPlus V3.5and Microsoft Excel.
16It is possibleto restatea portfolio optimi~tion problemto produceoff-frontier portfolios. Theseare assetallocationsfor
points in risk-return spacethat are within the concaveregion defmedby the setof efficient points. They are portfolios with
variancegreaterthanthe minimum variancepointsfor the sameexpectedreturns. They were found by goal equality
calculationusing the sameconstraintsaswere usedfor minimum varianceoptimization. However,the equality risk condition
was setto a higher level than found on the efficient frontier. Non-linear optimizationwas usedfor this purposewhereas
quadraticoptimizationwas usedfor minimum varianceoptimization.

Page15 of 45
The plug-in featuresof a bootstrapenableinferencefrom samplepropertiesof the distribution of [Link] plug-

in propertiesextendto all complex functionsof the bootstrap,including standarddeviations,means,medians,confidence

intervalsand any other measurablefunction. The EF is oneof thesefunctions.

The bootstrap is used in this paper to illustrate the impact of sampling error on the EF .17 EF is a complex function of the

historical returnsfrom which it was calculated. If the sampleis from a larger, unknowndomain,the bootstrapprinciplesapply.

In the caseof correlatedinvestmentreturns,a segmentof history might be thought of as a sample,but it may not be

operationallymeaningfulbecauseof samplingerror. Yet, tl¥: useof the historical datain DFA applicationstreatsit as though

it were both meaningful, representative and not a sample.

The behaviorof the EFs for our bootstrapsamplesis a non-parametrictechniqueusedto evaluatethe effect of samplingerror.

were history to be properly thought of as a sample. Becauseactuarialscienceis built largely on the preceptthat pasthistory,

evenof seeminglyunique phenomena,really is a sample,we proceedalong this slippery slopetoo.

3.2.1 Bootstrapping n- Tuples

The n-tuple observationof correlatedobservationsat time t canbe sampledwith [Link] techniquewas usedby

Laster [1998]. The experimentis similar to drawing packagesof colored gum dropsfrom a productionlot. Eachpackage

containsa mixture of different colors that are laid out by machineryin somecorrelatedmanner. Supposethe lot that hasbeen

sampledoff the productionline containsn packages.A bootstrapsampleof the lot also containsn [Link] is obtained

by [Link] replacement,from the original samplelot. The n-tuple of investmentreturnsat time t is analogousto a package

within the lot of gum drop [Link] historical sequenceof correlatedreturnsis analogousto the mix of different colors of

gum dropsin a package. The analogyhaltsbecausewe know the lot of gum drop packagesis a sample. We neverwill know

whetherthe sequenceof historical, n-tuple-investmentreturnsis a samplein a meaningfulsense.

The dataconsistof a matrix of monthly returns;eachrow is an n-tuple of the returnsduring a common interval of time for the

component assets (columns of the matrix); the value of n was ten and measures thc use of the ten investment categories

describedin Appendix A: Review of Data [Link] bootstrapmethodinvolves samplingrows of the original datamatrix.

Page16 of 45
An n-tuple describingthe actualreturnsfor assetcomponentsat an interval of time is drawn and recordedas an "observation'

in the bootstrapsample. Becausethis n-tuple can appearin anotherdraw, the processinvolves samplingwith replacement.

This randomizedchoiceof an n-tuple is repeatedfor eachobservationin the original sample. When the original samplehas

beenreplacedby a replacementsamplingof the sample,the result is referredto as a bootstrapsample. This processof drawing

a bootstrapsamplecanbe repeatedmany times, usually in excessof 2,000.

Eachbootstrapsamplehasboth a measurablecovariancematrix and an efficient frontier that can be derived using that

covariancematrix. It is unlikely that any two bootstrapsampleswill necessarilyhavethe samecovariancematrix. Each

samplecan be subjectedto mathematicaloptimizationto producean efficient frontier. The study askswhetherthis frontier is

stableacrossthe samples. Instability is measuredin two ways. First, the bootstrappedefficient frontier may fluctuate from

sampleto sample. This meansthat the distribution of risk for a return point on the EF is not a degeneratedistribution that

collapsesto a single point. Rather,there is a rangeof different portfolio risks amongthe bootstrapsamplesat a given return.

Thereis a probability distribution associatedwith risk, given a return amongthe bootstrapsamples. In other words, the study

attemptsto measurethe distribution, and the studyviews that distribution as a measureof samplingerror in risk-return spaceas

it impactson the calculationof an efficient frontier.

Second,the portfolio allocationsmay diverge qualitatively amongbootstraps. Wereportfolio allocationsto be aboutthe same

in an arbitrnrily small region of risk-return spaceamongdifferent bootstrapsamples,the practical effectsof samplingem>f

would be small.

3.2.1 Extension of the Bootstrap Sample as a DFA Scenario

The bootstrapsamplescan be usedin the way a DFA model might haveusedthe original historical data,including their direct

usewithin the calculationof the DFA resultsasa rnndominstanceof investmentresults. They are the sourceof DFA

[Link] papersuggestshow that direct useof the bootstrapmight unfold in a DFA liability-side simulation,but it does

]7 The bootstraphasbeenusedin connectionwith mean-varianceoptimi~tion by Michaud and othersin an attemptto improve

performance of EF portfolios. See Michaud [1998].

Page 17 of 45
not deploy it in that manner.18.19 The authors have a less ambitious objective of examining just the performance of the efficient

frontier built from bootstrappinginvestmentinformation.

3.3 Sampling Error within Risk-return Space

Thereis no clear-cutmethodfor estimatingsamplingerror that may exist in risk-return space. We do not know the underlying

distribution generatingan historical sample. We do not know whethera populationdistribution, were it to [Link],is stationary

over any time segment. We might, however,view history as an experimentalsample,particularly if we want to useit to

forecastcorporatestrategicdecisionsusing DFA

Samplingerror can be envisionedand approximatedin different ways for this hypotheticalunfolding of reality. Oneway is to

breakthe actualtime seriesinto arbitrary time segmentsand ask whethera randomselectionamongthe subsetsof time leadsto

different, operationallydisparateresults-these would be EFs basedon the sub-segmentof time that haveportfolio allocations

disparateenoughto be viewed as operationallydissimilar. If they are dissimilar enoughto warrant different treatment,a

samplingdistribution of interestis the one measuredby the effectsof thesetime-periodslices.

Anotherapproachis to envision prior history as an instantiation,period-to-period,from an unknown multivariatedistribution.

The samplingerror in this processis driven by a multivariatedistribution. Dependingon our model, we mayor may not place

dependencies
from prior realizationson this period's realization. That is, for DFA investmentreturn generationand intra-

periodportfolio rebalancing,the multivariate model may be stationaryor non-stationarywith respectto time.

3.3. 1 Michaud's Efficient Frontier

Michaud [1998] approachesthe measurementof samplingerror effectson EF in a different way. Although his approach

differs, his overall conclusionsare importantand consistentwith manyof our findings. He notes [1998, p. 33], "The operative

18Although the n-tuplc usedin this paperis a cross-sectionalobservationof returns,it canbe expandedto a cross-sectionof the
entirebusinessenvironmentat time t. This includesall economicaggregates,not just ratesof return. Any flow or stock
businessaggregatethat canbe measuredfor interval t is a candidatefor the n-tuple. This would include, inflation, gross
domesticproductor any worldly observationof the businessclimate prevailing at that time. A bootstrapsamplecanbe usedas
a componentof a larger simulationrequiring simulationof theseworldly events.
19DFA modelbuilders spendtime modelingempirical estimatesof processand parameterrisk [Kirschnerand Scheel,1998].
Bootstrappingfrom the dataremovesmuch of this estimationwork and leavesthe datato speakfor themselves.

Page18of 45
question is not whether MY optimi~tions are unstable or unintuitive, but rather, how serious is the problem. Unfortunately for

many investment applications, it is very serious indeed." Our paper will draw a similar conclusion.

He does not refer to an efficient surface but calculates a "resampled" portfolio that seems to capture some similar properties.

Michaud uses multivariate DOnna! simulations from the same covariance matrix used to calculate EFs. This covariance matrix

is from a sampleof data-the dataobservedduring somehistorical period. Justwhat definition of samplingerror basbeen

accommodatedin the Michaud resampledportfolio is unclear.

One of the Michaud simulationsis not equivalentto a bootstrapsampleusedin this study. Michaud's approachdoesnot

attemptto adjustfor a primary sourceof samplingerror-sampling error in the covariancematrix. In our study,eachbootstrap

samplehasan independentlymeasuredcovariancematrix. Using the DFAjargon of Kirschner and Scheel[1998], Michaud's

approachmay not accountfor parnmeterrisk in the [Link] ranking mechanismusedby

Michaud to combineEFs derivedfrom various multivariatenonnal simulationsmay distort risk/return spacebecauseeachEF

is segmentedin somenon-linearfashion to identify equally rankedpoints in risk/return space[Michaud, 1998,p. 46, footnote

11]. The portfolio profiles for identically rankedEF points [Link] it is not clear that equi-rankedpoints fall within

the same definition of risk/return space.

3.4/mportance to DFA Scenario Generation

This papercannotand doesnot attemptto rationalizethe processunderlying investmentyields over time ,20 Rather,the model

builder shouldbe careful to designthe DFA model to be in accordancewith perceptionsabouthow a samplingmethodology

may apply. The useof the model will invariably mimic that viewpoint

If, for example,oneviews history in the fashion imaginedby a bootstrapof n-tuples,and if that view doesobserveoperational

differences,then one can createscenariosfrom [Link] more theory is required. Hypotheticalinvestment

returnsarejust a bootstrapsampleof actualhistory.

20 What if therewere no commonobservablestationaryprobability measurefor securityprices? Kane [1999,p 174] argueswe

must use utility measurements.

Page19of 45
Similarly, ifEFs for historical periods produce superior performance in forecasting (compared to portfolios constructed from

off-frontier portfolios derived from the same data), then the use of an empirically detennined covariance model and

multivariate nonna! simulation makes a great deal of sense.

3.5/mporlance to DFA Optimization

Optimization often is used within DF A and cash flow testing models to guide portfolio rebalancing. The DF A model usually

grinds through the process of business scenario and liability scenario simulations before the optimizer is deployed. But,

accounting within the model often is done while the optimizer seeks a feasible solution.

The sequence of model events runs like this:

Independentlymodel many instancesof exogenousstatesof the businessworld (e.g.,assetreturns,inflation. measures

of economicactivity, monetaryconversionrates). Nwnber theseinstances,B" ~, B3,...,[Link] that eachof these

instances is a vector containing period-specific values for each operating fiscal period in the analysis.

Model many instancesof the company'sperfonnance. NumbertheseinstancesC\, ~ ,Cn.C\ often is dependenton

B. becauseit may usean economicaggregatesuchas inflation or economicproductivity to influence C1'sbusiness

growth or lossand expenseinflation. Each C is a vector spanningthe samefiscal periodsasB.

Observethat in someDFA modelsneitherB nor C is necessarilyscaledto the actualvolume of [Link] are

unit ratesof changefor underlyingvolumesthat areyet to be applied.

4. Let the optimizer searchmechanismposit a vector of weights that distributethe volume of assetsat to ' the inception

point for a forecastperiod.

s.
Apply the accountingmechanismsuscdby the DFA model to beginningassetsand accountfor the unit activities

expressedinB and c!\ Do this accountingfor eachvector pair {BI,CI}, {~,Cl},...,{BnCn} over the rnngeof its time

span.22

6. Calculatethe metric usedfor the goal and any constraintsasof the end of the fiscal period if it is a metric suchas

economicvalue or surplus. If it is a flow-basedmetric suchas portfolio duration or discountedGAAP income,derive

21At this stage,the derivation of taxeswould occur. As notedby Rowland and Conde[19%], the detemrinationof federal
incometaxesis convolutedby the combinedeffect of discountrates,changesin loss reserves,varying wlderwriting results,and
tax canyforwardsand canybacks.

Page20 of 45
 the metric for the holding period results. This calculation is done for each business/company scenario pair. There are

n resnlts; collectively they constitute a simulated sample.23

7. Return the required metrics for the sample to the optimizer. If the optimizer is deployed for EF calculation, the goal

will be a samplestatisticfor risk suchasvariance,semi-variance,or chance-constrained

percentileor range. The

sampleaveragefor the distribution developedin step(6) for the metric will be usedwithin the constraintset.

8. The optimizer will repeatsteps(4)-(7) Wltil it hasobtaineda feasibleset.

The optimizer usesa sample. The optimizer resultshavesamplingerror. Steps(1) and (2) are [Link] dterebe 10

repetitionsof this experiment. Application of steps(1)-(8) will result in 10 efficient frontiers, eachderivedfrom a different

experimentalsample. It is likely that they will havedifferent characteristics.

In a DFA experimentthere are manydrawsfrom the urn; eachsimulationis anotherdraw. The modelergetsdistributional

infonnation aboutthe contentsof the urn by the experimentalgroupingof all the simulations. When enoughsimulations

within each experiment are nm, convergence of the distribution of results can be achieved. Since it is unlikely for the output

distribution to be known, or necessarily capable of being parnmeterized, no a priori estimate is available. Instead, an empirical

measureof convergencemust be used.

The allocation of company assets among competing investment alternatives using a single efficient frontier calculation (based

on a single experimental result) may seem to be similar to betting on the allocation among balls of different colors within the

urn based on a single sample from the urn containing them. One may, or may not, be lucky. But, you improve your luck by

increasingthe numberof simulations.

One still may become victimized by a faulty decision while ignoring sampling error. This may arise in calibrating a model to

history. The historical record is a single draw from a true underlying probability distribution. We may be lucky that the

number of periods in the historical realization contains sufficient infonnation about the underlying process for unfettered

decision-making. But, we could be victims of sampling error, which we are unable to control or even limit

22 Somemodelsmay achievecomputationalefficiencieswhen economicscenariosare pairedwith E(C) insteadof with direct

pairing to C\, C2, , Cu. Whenthis is done,however,the varianceof the metric being optimizedwill be reduced,and the
minimum varianceportfolio is likely to be different.

Page21 of 45
4.0 Historical PerformanceComparison
Figure 4 illustratesthe performanceof severalportfolios over increasinglylonger forecastperiods. It showsresultsfor
24 The multipliers shownin the legendof Figure 4
portfolios, which, a priori, havedifferent levels of risk for the samereturn.

are multiples of the minimum variancerisk. The line for Multiplier-l tIacesthe performanceof the on-frontier, EF, portfolio.

Other lines in the figure with multipliers> 1 showperfonnanceof portfolios with the sameexpectedreturn,but higher

variance.

Figure4 Comparisonof Performancefor On-Frontierand Off-Frontier Portfolios

Performance I

~
..A~
762
- Multiplier 1
562
,362
~
A ~

-cO-
Multiplier 1.25
Multiplier 1.5
162 - Multiplier 1.75
~ Multiplier 2
-.038

-.238

Performance Information Ratio

-
Historical Period: January, 1988 December, 1992
-
Forecast:January,1993 December, 1999
Expected annualized return=.O825

23 If enoughpairs are used,the chancethat the model will convergeimproves.

24Risk in this study is measured as the standard deviation of return.

Page22 of 45
 25 It is known as the infonnation ratio. The
Figure 4 bares perfonnanceusing a variation of the Sharpeperfonnancemeasure.

Sharpeperformanceratio, which measuresexcessreturn to risk, is adjustedin the denominatorof the information ratio. The

denominatorof the Sharpeperformanceindicator is changedto excessrisk. The information ratio is given by (0.1):

E(rp-rf)
SD(rp-rf) (0.1)

where,

rp = monthly return on the portfolio,

rf = monthly return on the risk free componentof the portfolio26,

E = expectationoperator,

so = standard deviation operator.

Although the infonnation ratio was computedwith monthly data,it is expressedas an annualmeasurein the paper.

4.1 EF Performance Is Better for Low Risk-return Portfolios

The off-frontier portfolios, so-calledinefficient portfolios, achieveperfonnancethat rivals or bettersthat of the EF portfolio.27

Thereis no conceptof "significance" that can be attachedto the observeddifferences. However,it is clear that the

performancedifferencesare greatand [Link] portfolios out-performthe efficient one in the Figure 4. When

perfonnanceis measuredby geometricreturn. the under-perfonnanceof the EF portfolio can be more than 100basispoints as

shownin Figure 5. The underperfonnanceshownin Figure 5 is measuredover a seven-yearholding period,and therewasno

25Laster [1998] createdvarious portfolios by combining two assetcomponents,domestic(representedby S&P 500) and
foreign (representedby Morgan StanleyEAFE). His bootstrapsamplesof thesetwo componentswere usedto calculate
portfolio variance,assmningvarious mixes. He did not separatehistorical and forecastperiods. [Link] measuredquantiles
from the bootstrapsamplesafter constructingportfolios. He concludedthat diversification into foreign equitiessubstantially
changedand improvedthe risk/return profiles.
26 The 90-dayTreasurebill index is usedas the proxy for the risk free return.
27 Shortholding periodshaveperformancemeasures calculatedwith few [Link] ordinal rnnkingsamongthe
different multipliers are volatile and shouldbe ignored. The first six monthly periodsare generallyignored in this paper.

Page23 of 45
portfolio rebalancingduring this time. Data for other time periodsand the useof interveningportfolio rebalancingmight

materiallyaffect this evidenceof underperformance.

Figure 5 Comparisonof GeometricReturnfor On-Frontierand Off-Frontier

Portfolios

Performance

,079
,074 -Multiplier 1
,069 -0- Multiplier 1.25
,064 -A- Multiplier 1.5

,059 ~ Multiplier 1.75

L
~ Multiplier 2
.054
049 I I ,--r-

1 21 41 61 81
Geometric return
Historical Period: January, 1988 - December, 1992
-
Forecast: January, 1993 December, 1999
Expected annualized return=.O825

The performancevaries considerablywith the level of return and historical period. For example,Figure 6 illustrates

perfonnancefor an earlier period and a lower expectedreturn level. Here, the EF portfolio, does,indeed,out-performtre ofI-

frontier portfolios for about ten years. Thereafter,it reversesand perfonnancefalls below off-frontier portfolios. The Figure

illustratesthat the contemplatedholding period for useof an EF shouldprobablynot be as long. The performancevariance

illustratedin Figure 6 is volatile; the differencesin perfOmlancein on- and off-frontier portfolios varies considerablywith the

choiceof historical startingpoint and length of the holding period.

Page24 of 45
Figure 6 EF Portfolio Performance at Low Risk-return Levels

Performance

1.606
-Multiplier 1
-a- Multiplier 1.25
1.106
-.- Multiplier 1.5
-0- Multiplier 1.75
.606
-.- Multiplier 2

106
1 51 101 151
Performance Information Ratio
HistoricalPeriod: January, 1980-December, 1984
-
Forecast: January, 1985 December, 1999
Expected annualized return=.0649

4.2 Overall Behavior of On-Frontier Portfolios for Information Ratio

The historical recordwas examinedfrom severalperspectivesto seewhetheran EF portfolio continuesto out perform ofJ-

frontier portfolios. Equi-retum portfolios were examined. Theseare portfolios whosereturnsare the same,but they have

higher risk. The forecastperiod immediatelyfollowing the end of the historical segmentwas examinedto detenninehow long

the on-frontier portfolio [Link] forecasthorizon extendedto the end of the data,December,

1999. Historical segmentsconsistof a 5-yearblock of 60 observations.

Severaladjustmentswere madefor this analysis. The first six-monthperiod was ignoredbecausethe ratio is highly volatile

and computedfrom few [Link] exb"emelow return levelsalso were removedfrom the analysisbecausehigher ones

shownin the table dominatedthem.2S

-
28 The extremelow risk-return observationsoccur below where the EF curve hasa positive first derivative. A portfolio with a
higher return for the samerisk canbe found abovethis changein the curve.

Page25 of 45
Table 1 showsthe relative behavior of the information ratio at the return level indicatedat the top of the column. Eachrow

block includesthe time for subsequentrow blocks. For example,the forecastbeginning January,1980coversthe period

endingDecember,1999. The interval of measurementis a month. All of the other blocks begin at a later point, but all forecast

periodsend in December,1999.29

Missing cells in Table I indicate that a feasiblesetwas not found at that return level for one or more of the on or off-frontier

portfolios. Therewere five portfolios with risk up to two times the risk of the on-frontier point.

Table 1 Information Ratio Behavior

Forecast Period Return Levels

Infonnation Ratio (forecast begins 1/1980) 0.00661 0.008 0.0085 0.009 0.0095 0.01

:leriods until on-frontier point under perfonT1s(max=238)

Number of periods on-frontier point outperforms all others 10 14~ 148 153 154 151

Average on-frontier rank (5 is highest) 3.05 4.3~ 4.34 4.33 4.30 4.271

Information Ratio (forecast begins 1/1985) 0.0066 0.008 0.0085 0.009 0.0095 0.011

'jeriods until on-frontier point under performs (max=178) 11 110 109 119 69

Number of periodson-frontierpoint outperformsall others 105 104 10~ 11~ 124

Average on-frontier rank (5 is highest) 3.4~ 3.40 3.38 1.92 3.72 3.90

Information Ratio (forecast begins 1/1990) 0.0066 [Link]~ 0.0085 0.009 0.0095 0.01

~eriodsuntil on-frontier point under performs (max=118) 40

Number periods on-frontier point outperforms all others 34 66 83 103 11

Average on-frontier rank (5 is highest) 4.18 4.05 4.57 4.72 4.89 4.96

Infomlation Ratio (forecast begins 1/1993) 0.0066 0.008 0.0085 0.009 0.0095 0.01

"'eriods until on-frontier point under performs (max=82) 19 10

29Each block of rows usesa different set of on and off-frontier portfolios-the respectiveEFs are derivedfrom optimizations
on differentperiods. For example,the January,1980forecastis basedon the perfonnanceof EFs derived from an historical
segmentcoveringthe five-year period, January,1975- December,1979). However,the January,1995forecastusesEFs
derivedfrom a different period, one covering the five-year ~riod, January,1989- December,1994. The information in the
blocks is not cumulative;the nwnber of periodsthe on-frontier excelsor outperformsoff-frontier portfolios is a separate
measurementfor eachrow block. The row blocks show performancefor portfolios constructedat different points in time.

Page26 of 45
 Forecast Period Return Levels

~umber of periodson-frontierpoint outperformsall others 15

IAverage on-frontier rank (5 is highest) 2.10 1.83 1.8~ 1.81! 1.60 1.56

Information Ratio (forecast begins 1/1995) 0.00661 0.008 0.0085 0.009

:)eriods until on-frontier point under perfonns (max=58) 5~ 56 57 Never

Number of periods on-frontier point outperforms all others 4~ 50 5" 53

Average on-frontier rank (5 is highest) 4.8~ 4.94; 4.96 5.00

"Periodsuntil on-frontier point under perfonns" meansthe first period that an off-frontier portfolio beatsthe on-frontier

efficient portfolio. "Number of periods on-frontier outperformsall others" meansthe last period where the efficient portfolio

wins. Performancetendsto hold up better for lower return levels. This effect is reinforcedby the larger valuesshownfor the

numberof periodsthe on-frontier portfolio doesout rank the off-frontier portfolios. In genernl,the on-frontier portfolio ranks

well comparedto the others. The averagerank is generallyhigh, above3 out of 5. But. the performanceis not [Link]

on-frontier portfolio did well during the long forecastperiod startingJanuary,1980and during the shorterforecastperiod

startingJanuary,1995. However,the low averageof the on-frontier for the January,1993showsthat the perfonnanceis

greatly influencedby the historical period and perhapsinfluencedby samplingerror.

Therealso is greatinconSistencyin the nwnber of periodsbeforean off-frontier portfolio hasa higher infonnation ratio. The

scanbeginsin period 6 of the forecasthorizon. so the reversalshownin the tablewill either be neveror a numberbetween6

and n. In most cases,the reversalis early, but not pennanent Thereare many situationswherethe on-frontier portfolio wavers

betweenhighestrank and somethingless. This latter fact is found in the rows, "Number periodson-frontier outperfonns. In

most casesthis numberis larger than the numberof periodsbefore [Link] that the on-frontier waflles in and out

of [Link] could be anotherindication of samplingerror. The choiceof an on-frontier point may no~ and

probablydoesnot, imply superiorperfonnance.

4.3 Behavior for Other Performance Measures

The information ratio is believedto be a valid measureof perfonnancebecauseit adjustsfor variation in the return series

during the period of measurement Were it applied to two consultants'portfolio allocationrecommendations,the consultant

Page27 of 45
with lower excessreturnscould be rankedhigher than the other consultantbecauseof proportionatelylower risk in excess

return. This may be small consolationto the bolder of the lower wealth portfolio recommendedby the higher ranked

consultant. This is why it is important to assessother characteristicsbeyondthe appetitefor risk before making an allocation

decision. The managerwith the higher infonnation ratio hasthe better cost of risk per unit of return; yet. it is not of muchuse

if a minimum return level or endingwealth is required.

Thereis considerablehistoric instability in the standarddeviationof returns. This can be seenin Figure 7, which showsthe

historic progressionof changesin the standarddeviationof monthly returnsof the portfolio componentsusedin this study.
The lines showthe changein standarddeviationfor rolling five-yearblocks of data.30 Any performancemeasuretl1atis a

function of this risk proxy, suchas the infonnation index, will be inherentlysensitiveto suchvolatility and, perhaps,exhibit

similar historic instability. This volatiljty m risk helpsto explmn why WstoricalEFs may lack forecastpower.

30 Therewas significant volatility in the securitiesmarketsin 10/87("Black Monday") and 8/98 (Long Term Capital crisis).
These periods are highlighted in the figure.

Page 28 of 45
 -- T-t: I ~I
~
Q)
~
~
Q)
c
[J
-.., \»"
)
N
I
w
~ t it
~
~
"a5
~.
. I"
A(
.. $ .0T
~
~
.
-41+; ~
,1 ~
~
..,
,
r
8
~ ~
.!- .
!l -
i~~ .
II)
c
0
0
N
~
.m
C
>
Q) -
-~ "'I
"C J ~
"0
...
( -~-
0-
~
"C
\~ N
-
C
~ \. ~
p.,
it'
.--.
(/)
>- i
:E
C
0
~
c
.
...
~ ~
0 N
~ Q)
>- .** *\ \ ~
ca
Q)
>
Q)
> .0-
ii: 0'
C
0) ) Il)
,
D..
"C
.5 ~
-
'-
ca (/)
"0
"C
C
ca
...

~
~ I
0
cn
c
'-
~ + +
-
G
0
«
m
-.J
...
Q) .
ft:
C
c po
~ ~
:I:
!
~
...
~
ca
0
-
N
~
.
>
to..
Q)
1. - §
'-
~
C)
~
,...
~
0
<0
~
uo) .~ ~
coj
~
N ~
..- ~
0"
.
u..
One measureofperfonnance that is not risk-adjustedis geometricreturn during a holding period. Resultsareanayedin Table

2. The layout of this table is similar to Table 1

Table 2 Geometric Return Behavior

Forecast Period Return Levels

Geometric Return (forecast begins 1/1980 0.0066 -0.0081 0.0085 0.009 0.0095 0.01

~eliods until on-frontier point under performs

(max=239)

Number of periodson-frontierpoint outperforms 107 106 102 102 105

~verage on-frontier rank (5 is highest) 2.4 3.35 3.39 3.3~ 3.38 3.40

Geometric Return (forecast begins 1/1985 0.0066 0.0081 0.0085 [Link]~ 0.0095 0.01

~eriods until on-frontier point under performs

l(max=179)
Number of periodson-frontierpoint outperforms

fAverage on-frontier rank (5 is highest) 1.00 1.00 1.00 1.00 1.03 1.04

Geometric Return (forecast begins 1/1990) 0.0066 [Link] 0.0085 [Link]~ 0.0095 0.01

~eriodsuntil on-frontier point under performs

(max=119) 21 74 89 11 119

Number of periods on-frontier point outperforms 15 68 85 105 113

Average on-frontierrank (5 is highest) 4.1 4.06 4.60 4.75 4.92 4.9~

Geometric Return (forecast begins 1/1993) O.OO6~ [Link] 0.008~ [Link]~O:OO95 0.011

:leriods until on-frontier point under pelforrns (max=83) 15 15 16 16 16 16

Number of periodson-frontierpoint outperforms 1 16 18 1~ 1- 10

Average on-frontier rank (5 is highest) 1.99 2.1~ 2.22 2.22 1.92 1.73

Geometric Return (forecast begins 1/1995) 0.0066 [Link]~ 0.0085 0.009

':'eriods until on-frontier point under performs (max=59)

Page30of 45
 Forecast Period Return Levels

~umber of periodson-frontierpoint outperforms 30

fAverage on-frontier rank (5 is highest) 2.44 2.80 2.96 3.3~

The forecastpropensityof the on-frontier allocation is markedlychanged. Wealthgrowth appearsto be unrelatedto the on or

off-frontier portfolio choice,and often is worsefor the on-frontier allocation. The numberof holding periodsthe efficient

frontier portfolio dominatesoff-frontier portfolios is generallya lower proportion of the possiblenwnber of holding periodsin

Table 2 than in Table Michaud [1998, pp. 27-29] claims thereis a portfolio within the EF, the "critical point," below which

single period mean-varianceefficient portfolios are also n-periodgeometricmeanefficient and abovewhich single periodMY

efficient portfolios are not n-period geometricmeanefficient

Performance Failure within CAPM

Work with betahasled to various criticisms [Malkiel, pp. 271].3] For example,somelow risk stocksearnhigher returnsthan

theory would predict Other attackson betatend to mirror what we seewith EF:

.
Capital assetpricing model predictsrisk-free ratesthat do not measureup in practice.

2. Beta is unstableand its value changesover time.»

), Estimatedbetasare unreliable:4

4. Betas differ according to the market proxy they are measured against.35
Averagemonthly return for low and high betasdiffers from predictionsover a wide historical span.36
5.

31 Beta is a measureof systematicrisk either for an individual securityor for a portfolio. High betaportfolios, measuredex
ante,in theory shouldhavehigher returnsexpost than low beta portfolios.
32Whenten groupsof securities,rangingfrom high to low betas,were examinedfor the time period 1931-65,the theoretical
risk free rate predictedby CAPM and actualrisk free ratessignificantly diverged. Low-risk stocksearnedmore and high risk
stocksearnedlessthan theory predicted. [Malkiel, pp. 256-7]
33 During shortperiodsof time, risk and return may be negativelyrelated. During 1957-65,securitieswith higher risk

~roducedlower returnsthan low beta securities.[MaIkiel, pp. 258-60]

The relationshipbetweenbetaand return is essentiallyflat. Beta is not a good measureof the relationshipbetweenrisk and
return. [MaIkiel, pp. 267-8]
35 Predictionsbasedon CAPM aboutexpectedreturnsboth for individual stocksand for portfolios differ dependingon the

chosenmarketproxy. In effect, the CAPM approachis not operationalbecausethe true market proxy is unknown. [MaIkiel,
pp. 266-7]

Page31 of 45
Malkiel [po270] concludesfrom his surveythat, "One's conclusionsaboutthe capital-assetpricing model and the usefulnessof

betaasa measureof risk dependvery much on how you measurebeta." This appearsto be true of EFs too. The definition of

efficiency is what is important here-perbaps more importantbecausecorrect measurementrequiresprecisedefinition.

The choiceof an optimization mechanismcouchedin tenDSof risk-returntrade-off may not lead to wealth maximization.

Under thesepretensesone might wish to deploy a different optimization mechanismsuchas the onementionedby Mulvey, et

oJ [1999,p. 153] in which the optimiZ3uonseeksto maximizeutility. The choiceora particular utility function may be framed
in tenns of absoluterisk aversion-negative exponentialutility works in this regard37 And. if the behaviorof securityprices

doesnot havean observablestationaryprobability measure[Kane, 1999],utility approachesseemto be mandatory

The subjectof what is optimal is controversiaLand not apt to go away. The useof optimization within hybrid modelsand

generationof metricsby DFA modelshasmany [Link] is the choiceof planning horizon. Michaud [1998,

p. 29] arguesthat investorswith long-tenn investmentobjectivescan avoid possiblenegativelong-tenDconsequences

of mean-

varianceefficiency by limiting considerationto EF portfolios at or below somecritical point. There is a parallel in our paper,

in what we refer to as samplingerror and its affect on the shapeof the efficient surface. This surfaceappearsto haveproperties

at the lower risk-return areasof both lower dispersion,greatersimilarity in portfolio composition,and betteron-frontier

performanceamongdifferent samples(eitherbootstrapor historic segment).

5.0 Characteristics of the EF Surface

The bootstrap-generated
EF surfaceriseswithin the risk-return space. Views of this surfacefrom two different anglesarc

shown in Figures 8.

36 Theratioof priceto bookvalueandmarketcapitalization

did a betterjob of predictingthestructureof nonfinancial
corporatesharereturnsthan beta during a 40-yearperiod. EugeneF. Famaand Kcnncth R. French,"The Cross-Sectionof
ExpectedStockReturns,"Journal of Finance, June, 1992.
37The recommendationof a utility-decision approachhasgreatbreadthin the insuranceliterature-beyond the useof utility as
goal function in optimization, other venuesfind it appropriatewhere stochasticdominanceis sought. For example,exponential
utility usewas suggestedin rate making by Freifelder. SeeFreifelderLeonardR, A Decision TheoreticApproach to Insurance

Page32of 45
Figures 8 Views of EF Surface Createdfrom Bootstrap Samples

The surfaceis constructedfrom monthly returns. Looking down on the surfaceof the views, one obtainsa projectionon risk-

return space. The surfaceis seento curve as the efficient frontier curves. In the low risk-return sector,the surfaceis more

Ratemaking,hwin, 1976,pp. 141. The choiceof parametersfor utility functions is perhapsas much an art as the

Page33of 45
peaked. The surfaceflattens and broadensin the risk-return space. Imagineyourself walking along the ridge startingin the

southwestand proceedingnorthwardand then northeast. You would first be descendinga steepincline and then a vista of a

vast planewould unfold along your right. This canbe interpretedwithin the contextof changesin the marginaldistributions

representingslicesthrough the surfaceeither along the risk or along the return dimensions. We refer to the latter asan equi-

return slice, and its propertiesare examinedin more detail at a latter point in the paper. In either case,the visualizationis one

of moving from lessdispersedmarginal distributionsto oneswith greatervarianceas either dimensionis increased

Thereis an artifact of the intervalization that resultsin a suddenrise in the surfaceat the highestrisk level. This occurs

becausehigher risk observationswere lumpedinto this final interval. Were higher levelsof risk intervalizedover a broader

rnnge,this ridge would flatten.

The surfaceshownin either of the views in Figures8 is built from manyefficient frontiers, eachproducedfrom optimizations

doneon a bootstrapsample. We alreadyhaveseenin Figure I a subsetofEFs that tangletogether-they can be organizedto

producea surface. The surfacedevelopsthe sameway an empirical probability distribution is built from a sample. Repeated

samplingproducespoints that are intervalizedand counted

A frequencycount can be madeof observationsfor EFs falling within an arbitrnrily small, two-dimensionalregion of risk-

return space. An exampleof this mappingfor 5,000bootstrap-simulated

EFs appearsin Figure 8. Collectively, this mapping

involvesthe 2-dimensional,intervalizationof approximately45,000quadraticoptimizationsconstitutingthe EFs for the

underlying bootstrapped samples.38

5.1 Equi-Return Slice of the Efficient Surface

A slice throughthe efficient surfacealong the returnplaneproducesa histogramof the minimum risk points for a given return

in the EFsusedfor the EF Surface. As return increases,this marginalprobability distribution becomesmore disperse. An

exampleappearsin Figure 9.

~eterization of claims generations in DF A models.

Equi-return, minimwn variance points for the 5,000 bootstrapped EFs were intervalized based on an overall evaluation of the
range of risk among all points on all EFs. If an efficient set could not be identified for a return level, the observation was
ignored. The marginal probabilities (risk-return) were normalized to the nwnber of viable observations for that risk level. The
nwnber of viable optimizations exceeded 4,500 at each return level.

Page34of 45
 Figure 9 Dispersion of Risk Given a Return Level

Distribution of Risk

0.25
>- 0.2
~

.
:s 0.15
.g 0.1
Q: 0.05

0
M M M "I" "1"10 10 10 <0 <0
~lOo)M""'~lOo)M"'"
~~~NNMMM"I""I"
0000000000
0000000000
Risk at Expected Return-.OO9

The dispersion increases with return for both surfaces constructed from bootstrap samples and from randomly selected blocks

of history. The distributions are positively skewed. increasingly so as return increases. The inset bars in Figure 9 identify the

intervals containing tile mean and median points of tile distnbutioo. Additional statistics botil for bootstrapped and historical

segment evaluations of sampling error appear in Tables 3 and 4.

Page35of 45

~
Table 3 Statistics for Equi-Return Slices of the Efficient Surface

Statistic Efficient Surface from Bootstrapped Efficient Frontiers

ReturnLevel 0053 0066 0080 .0085 0090 0095 0100

Mean .0125 0533 3.76 8.86 17.9 33.6 50.7

(times 1.0E4)

Standard .627 3.77 38.2 58.6 81.8 109.7 131.6

Deviation

(times 1.0E4)

Skewness 000123 378 57.8 136 27S 516 779

(times [Link])

The statisticsarevisually apparentin the EF surfaceshownin Figures8. The surfaceis partially bowl-like-sloping

downwardin a concavefashion. Its rim encompasses

a planewithin the risk-returndomainthat is broad in the risk dimension.

As one movesfrom low to high return, the marginaldistribution ofEF points measuringoptimized risk (an equi-returnslice

throughthe surfaceas illustrated in Figure 9) becomesmore [Link] a visual contextas one movesfrom low to high risk

along the EF surfaceand takesequi-returnslicesthrough it, one would find highervariancein the distribution of optimizedEF

risk points-variance shownin histogramplots suchasFigure 9 is greater.

An efficient surfacealso can be createdfrom EFs calculatedfor historical time periods. An exampleappearsin Figure 10. The

dataarefor 5-year,overlappingblocks calculatedon a monthly basisstarting in 1970. The samegeneralfeaturesarefound in

this representationof sampleerror. However,the surfaceis lessflat than the one developedfrom [Link]

reduceddispersionin the surfaceof Figure 10 arisesin part from the useof overlappingfive-year blocks usedto constructthe

underlyingEFs from which the surfaceis built. A statisticaltable similar to Table 3 was constructedfor this surface. It

appears in Table 4.

Page36of 45
Figure 10 Efficient Sutface from HistoricalSamples

Bflclent Surface
(based on historical segments)

Table 4 Statistics for Equi-Return Slicesof the Surface Shown in Figure 10

Statistic Efficient Surface from Bootstrapped Efficient Frontiers

ReturnLevel 0053 0066 ,0080 0085 0090 0095 0100

Mean 663 5.53 21.0 25.5 28.8 33.2 46.5

(times 1.OE4)

Standard 080 451 861 940 995 1.06 1.23

Deviation

(times 1.OE2)

Skewness .00676 769 2.92 3.54 4.00 4.62 6.46

(times 1.0E6)

Page37 of 45
 6.0Stability ofPol1folio Composition Along an Efficient Frontier
Portfolio allocation amongcomponentsecuritieschanges,usually dramatically,along the efficient frontier. A componentmay

enterthe feasibleset at somepoint, increasein weight, decreaseand then drop out at anotherpoint along the EF. This effect

was shownin Figure 2.

The change in composition for an equi-retum level was examined among different EFs, constructed both from historical

segment EFs and bootstrap EFs. We refer to this type of comparison as an avalanche chart because when shown in an

animation, the change in composition is similar to an avalanche. An example appears in Figure II

Figure 11 Avalanche Chart for Historical Segments

The vertical barsare stackedcolumns. Each segmentwithin a column representsa different componentof the portfolio. A bar,

therefore,comparesthe percentagevalue eachcomponentin the feasibleset contributesacrossall componentsin the set. All

barsare shownfor a constant,equi-returnlevel of an EF; but eachbar is for a different historical segment In Figure 11, each

bar representsthe portfolio compositionfor the equi-returnlevel point on the EF, which was caicuiatcdfor a five-year block of

monthly observations. The barsare for ten randomlychosenhistorical segments.39

Were the blocks within the barsto consist

39 There is a small chance that two or more bars in an avalanche chart could be identical. However, there is a much larger

probability that two or morebars have overlapping time periods in the calculation of their respective EFs.

Page38 of 45
of the samecomponentsand were they to be about the samesize, the portfolio allocationswould be the sameregardlessof the

time frame. Examinationof Figure 11 showsthat the compositionof the bars and individual componentallocationsvaries

considerably.

The portfolio compositionis much more stableat lower risk-return levels. This result is in accordancewith other similar

findings basedon the EF [Link]. too, showslessdisperseresultsfor lower return levels. This approachto measuring

samplingerror implies that performanceof efficient frontiers may not be optimal relative to off-frontier portfolios. If the mix

and compositionof portfolios fluctuatesconsiderablyboth with respectto historical aIKibootstrapsamplingmethods,the

performanceexpectationsof an ex anteallocation are not apt to hold expost.

7.0 Conclusion

Thebehaviorsshownin both Tables 1 and 2 illustrate a [Link] efficient surfacebuilt from

bootstrapsamplesis highly variable within the risk-return domain. Thereappearsto be sometemporaldominanceof on-

frontier portfolios for lower risk-return levels,but the historical recordis mixed. The bootstrappingof the single sampleof

assetreturnsprovidedby the historic dataillustratesthat samplingerror could materially affect the position and shapeof the

efficient frontier.

7.1 Should Efficient Frontiers Be Used in DFA Models?

Thereis no strongsupportin this paperfor the practicaldeploymentof efficient frontiers in DFA. The risk in DF A models

stemsfrom model,processand parameterrisk. It impactsthroughall aspectsofDF A modelsof the [Link]

existenceof model and processrisk [Kirschnerand Scheel,1997]thwartsthe usualconvergenceto the true w1derlying

distributionsgainedby running largenumbersof simulations. Whenall of thesenew risk elementsare heapedon top of the

samplingerror derivedfrom assetmodel calibration or empirically measuredcovariancematrices,one wonderswhetherEFs

are really useful in DFA analysis.

The work of Michaud [1998] bears on the issue of improving the perfonnance ofEF portfolios. He defines a measure of

statisticalequivalencefor mean-varianceefficiency. Any portfolio within the efficient surfacesufficiently closeto the optimal

portfolio is consideredequivalentto it. The extensionof his ideato the efficient frontier surfaceis to identify a region on it

Page39of 45
whoseex ante chance-constrained
probability both canbe measuredand hasdesirablestatisticalpropertiesin a forecasting

sense. This is analogousto acknowledgingthe existenceof samplingerror and specifyingan unknownpopulationparameter

only to within an interval of statisticalconfidence. Unfortunately,the definition of sufficiently closeis constructivebut

difficult to implementin a rigorous manner,particularly within the contextof the hybrid DFA model.

Future studywill haveto answerthe questionof whetheron-frontier assetallocationsthat are measuredfrom hybrid DFA

modelssuffer a similar unreliability. But, the problemswith on-frontier assetportfolios raisedin this paperare apt to be

exacerbatedby inclusion of known samplingerror in tile liability side of DFA models.

7.2 How Can EFs Be Efficiently Deployed?

Usersof this constructshouldbe awarethat the tenD"efficient" in efficient frontiers hasa good chanceof being operationally

false. The efficiency of portfolio compositionis unlikely to be manifestin better perfonnanceof the on-frontier portfolio

comparedto other, off-frontier portfolios. The risk/return surfaceis not adequatelymeasuredby a single EF, and sampling

error may lead to unwarrantedconclusionsaboutthe efficacy of portfolios measuredin suchsingularoptimizations.

The userofEFs shouldprobablyview them as containingprovisional,useful information about risk/return [Link],

any singleEF haslimited value in understandingthe risk/return surface. The conceptualbasisof an efficient surfaceis an

organizedresamplingof the dataso that the decisionprocessbenefitsfrom betterwlderstandingof uncertaintythat might arise

just becausethe EF is operationallyderivedfrom a sample. The misunderstandingof this uncertaintymay lead to erroneous

decisions,and the practitionermust be alert to potentialinefficienciesof a singleEF [Link] authorsrecommend

the elicitation of an efficient surfacebecausethe surfaceis apt to show a lack of statisticalconfidencein any single frontier on

that surface. Under thesecircumstances,the practitionermust think in tenDSof confidenceranges. The samplingerror shown

in the efficient surfaceemphasizeshow careful one mustbe when drawing inferencesderivedfrom optimization. An

optimizedfrontier is basedon an empirical covariancematrix; one that hassamplingerror. That error may be very important.

It is easyto believe that stmtegicor tactical decisionsmotivatedby so-call optimizedDFA measurementwill effectively move

the organizationto a betterposition in risk/return space. Unfortunately,thereappearsto be a broad region of "inefficiencY"

that may serveas well. An EF may be better than a crystal ball; but thereis a goodchancethat it should not be takentoo

seriously.

Page40 of 45
Appendix A: Review of Data Sources

This paperusesmonthly time seriesof assetclasstotal returns. A selectionof broad assetclassestypical of P&C insurance

companyassetportfolios was chosenfor examination. The time seriesall begin January1, 1970. However,certainasset

classes (e.g. mortgage backed securities) do Dot have a history that extends back this far. For these classes the time series were

backfilled to the January I, 1970 start date by an investment. consultant. The backfill process was based on a consideration of

the market conditions of the time (e.g. interest rates, fixed income spreads, inflation expectations) and how the particular sector

would have performed given those market conditions. The Start Date in Table 5 refers to the date historical data begins.

Table 5 Asset Components

Class Code Source Start Date

InternationalEquities EAFEU MSCI EAFE Index 1/1970

InternationalFixed Income INTLHDG JP Morgan Non-US TrndedIndex 1/1970

Large CapDomesticEquities S&P5 s&P 500 Index 1/1970

Cash usm 90 Day US TreasuryBill 1/1970

Mid CapDomesticEquities RMID S&P Mid Cap400 Index 1/1982

High Yield mYLD CSFBHigh Yield Bond Index 1/1986

ConvertibleSecurities CONY CSFB ConvertibleIndex 1/1982

CorporateBonds LBCORP LehmanBrothersCorporateBond Index 1/1973

GovernmentBonds LBOOVf Lehman Brothers Government Bond Index 1/1973

MortgageBackedSecurities LBMBS Lehman Brothers Mortgage Backed Securities Index 1/1986

Page41 of 4S
The time seriesusedin this study are monthly returns. With the exceptionof work relating to performance,all returnsare

expressed as monthly returns.

For perfonnancemeasurementpurposes,returnshavebeenannualizedusing the following formulas.

AnnualizedExpected Return

Rp =(I+T,)12-1 (0.2)

where,

R = annualized return,
p

rp = monthly return.

Annualized Varianceof Return

Vp =[vp +(1+ jlpt]12 -(I+MpY (0.3)

where,

annualizedvarianceof return,

monthly varianceof return,

expectedmonthly return.

M p = expectedannualizedreturn.

Annualized GeometricReturn
The growth rate,g, for a holding period of n yearsis given by:

Page42 of 45
1+ g =(~tl" (0.4)
Va

where,

v" = portfolio value at the end of the holding period 0,

v0 = portfolio value at the beginning of holding period.

Page43 of 45
References
Chopra,Vijay and Ziemba, William, "The Effect of Errors in Means,Variances,and Covarianceson Optimal Portfolio

Choice", WorldwideAsset and Liability Management,1998,CambridgeUniversity Press,pp. 53-61

Con-enti,Salvatore,Sontin,StephenM. and Isaac,Daniel B., "Applying a DFA Model To Improve StrategicBusiness

Decisions,"CasualtyActuarial SocietyForum, Summer1998,pp. 15-52.

Davison,AC. and Hinkley, D. V., BootstrapMethodsand Their Application, 1997,CambridgeUniversity Press,pp. 582.

Efron. GeraldF. and Tibshirani, Robert J.,An Introduction to the Bootstrap, 1993,Chapman& Hill, pp.436.

Elton, Edwin J. and Gruber,Martin J.,Modem Portfolio Theoryand InvestmentAnalysis, 1995,John Wiley & Sons,Inc., pp.

715.

Kane, S.A., "What happened to the utility functions? Imprecise expectations of security prices," International Review of

Financial A na/ysis,8:2 (1999),pp. 165-175.

Kirschner,GeraldS., "A Cost/BenefitAnalysisof Alternative InvestmentStrategiesUsing Dynamic Financial Analysis

Tools", CasualtyACbJarialSocietyForum, Summer2000,pp. 21-54.

Kirschner, Gerald S. and Scheel, William C., "The Mechanics of a Stochastic Financial Model," 1998 Proceedings of the

CasualtyActuarial Society,~ Nos. 162 and 163,pp. 404-408.

Laster,David S., "Measuring Gainsfrom InternationalEquity Diversification: The BootstrapApproach," TheJournal of

Investing,Fall, 1998,pp. 52-60.

Page44 of 45
Levy, Haim and Gunthorpe,Deborah,"Optimal InvestmentProportionsin SeniorSecuritiesand Equities Under Alternative

Holding Periods,"TheJournal of Portfolio Management,Summer1993,p. 31.

Malkiel, Burton G.,A Random WalkDown Wall Street, 1996,W. W. Norton & Company,Inc., pp. 227-276.

Michaud, Richard0., Efficient AssetManagement,1998,HarvardBusinessSchoolPress.

Michaud,Richard 0., "The Markowitz Optimization Enigma: is 'Optimized' Optimal", Financial AnalystsJournal,45, Jan.-

Feb.,pp 32-42.

Mulvey, JohnM, Madsen,Chris K. and Morin, Fran~is, "Linking StrategicandTactical Planning Systemsfor Dynamic

Financial Analysis," CasualtyActuarial SocietyForum, Summer1999,pp. 149-167.

Rowland, Vincent T. and Conde,Frank S., "Dynamic Financial AnalysisIssuesin InvestmentPortfolio Management,"

CasualtyActuarial SocietyForum, Spring 1996,pp. 205-40.

ShaIpe,William F., "The SharpeRatio", TheJournal of Portfolio Management,Fall 1994pp. 49-58

Page45 of 45