Performance Measurement

0.0(0)
Studied by 0 people
call kaiCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/69

encourage image

There's no tags or description

Looks like no tags are added yet.

70 Terms

1
New cards

4 attributes of an effective attribution process

  1. account for all the portfolio’s returns or risk exposure

  2. reflect the investment decision-making process

  3. quantify the active decision of the PM

  4. provide a complete understanding of the excess return/risk of the portfolio


2
New cards

Performance attribution

how that performance was achieved or how the risk was incurred

  • explain absolute or relative return

  • what portion was driven by active mgr. decisions

  • decompose excess return into component sources

  • decompose risk


3
New cards

Performance Evaluation

Draw conclusions regarding the quality of performance

Distinguish manager luck from skill

4
New cards

Return attribution

analyze the impact of the active investment decision on returns

5
New cards

Risk attribution

analyze the risk consequences of those decisions

absolute or benchmark relative terms

6
New cards

Micro attribution

understanding the drivers of a mangers returns and whether those drivers are consistent with the stated investment process

analyzes the small decisions that lead to gains/losses such as security selection, sector allocation

7
New cards

Macro attribution

Decision #1: deviations from the strategic asset allocation

  • success = overweighting an asset class that outperforms its benchmark (or combined weighted benchmark)

Decision #2: selection of investment managers

  • seccess = selecting a manager that outperforms their benchmark


8
New cards

Return-based attribution

Uses historical total portfolio returns to estimate the sources of performance without examining individual holdings.


  • Best used: When detailed portfolio holdings or transaction data are unavailable.

  • Advantage: Easiest and least data-intensive to implement.

  • Disadvantages: Least precise; relies on statistical estimates and is vulnerable to data manipulation or model selection bias.


9
New cards

Holdings-based attribution

uses actual holdings (beginning of period)

all transactions are assumed to occur at the end of the day

accuracy improves when data has shorter time intervals

most appropriate for investment strategies with little turnover

10
New cards

Transaction-based attribution

Uses both the holdings and the transactions during the evaluation period

Most accurate, but most difficult and time consuming to implement

  • more complex version of holdings-based


11
New cards

Top-down vs Bottom Up

  • Top-down: Start with the total return → work backward to estimate the sources.

  • Bottom-up: Start with individual securities/trades → build up to the total return.


Top-down → returns based

Bottom-up → holdings-based and transaction based

12
New cards

Brinson-Hood-Beebower (BHB) - Equity Atrribution

Use the chart like you are trying to find the area (length * width) of the square you need

Rewards mangers when they overweight a sector with return > 0 and underweight a sector with a return < 0

  • drawback is that if all sectors returns > 0 you can’t overweight them all so you are punished and on the alternative if all sector returns are < 0 you can underweight them all


To draw square AI is bull shit (BS)

<p>Use the chart like you are trying to find the area (length * width) of the square you need</p><p>Rewards mangers when they overweight a sector with return &gt; 0 and underweight a sector with a return &lt; 0</p><ul><li><p>drawback is that if all sectors returns &gt; 0 you can’t overweight them all so you are punished and on the alternative if all sector returns are &lt; 0 you can underweight them all</p></li></ul><p></p><p>To draw square AI is bull shit (BS)</p>
13
New cards

Brinson-Fachler (BF) - Equity Atrribution

Differs from BHB only by the second term (rb - Rb) versus just rb

rB -RB = return on the sector - return on the benchmark.


Therefore, it takes into account if the sector beat the benchmark, whereas BHB does not.

<p>Differs from BHB only by the second term (rb - Rb) versus just rb</p><p>rB -RB = return on the sector - return on the benchmark. </p><p></p><p>Therefore, it takes into account if the sector beat the benchmark, whereas BHB does not. </p>
14
New cards

Factor-based - Equity Attribution

Fundamental factor model

decompose contributions to excess returns from factors


RMRF = return on market minus risk free rate → market risk premium

SMB = small minus big → small-cap vs large-cap stocks

HML = high minus low → value vs. growth stocks (book-to-market ratio)

WML = Winners minus losers → momentum: past winners vs. past losers

<p>Fundamental factor model</p><p>decompose contributions to excess returns from factors</p><p></p><p>RMRF = return on market minus risk free rate → market risk premium</p><p>SMB = small minus big → small-cap vs large-cap stocks</p><p>HML = high minus low → value vs. growth stocks (book-to-market ratio)</p><p>WML = Winners minus losers → momentum: past winners vs. past losers</p>
15
New cards

Exposure Decomposition - Fixed Income Attribution

Top-down approach

Benchmark vs Portfolio

  • Benchmark

    • duration

    • yield curve positioning

    • sectors (i.e. government corporate)

  • Portfolio

    • active decisions to deviate from benchmark exposures


how well, relative to the benchmark, did active decisions work out, what contribution to active return

  • used primarily for client reports, easy to understand


Examples:

  • active duration bets → increase duration relative to benchmark in anticipation of falling rates

  • yield curve positioning → barbell for a flattening curve

  • sectors → overweight credits in anticipation of spreads narrowing


16
New cards

Yield curve decomposition (duration based) - Fixed Income Attribution

Can be either top-down or bottom-up

Estimates the returns based on duration

  • %total return = %income return + %price change

    • %price change = -ModDur * Δyield


Applied to both the portfolio and the benchmark

  • difference = effect of active PM decision

requires more data points than exposure decomp.

Typically used in reports for analysts and PMS


17
New cards

Yield curve decomposition (full repricing) - Fixed Income Attribution

Instead of using estimates

  • therefore does not have residuals unlike duration based

prices out each security

most complex attribution of the three fixed income approaches

18
New cards

Considerations in selecting a risk attribution approach

knowt flashcard image
19
New cards

Liability-based benchmarks

  • focus on the cash flows the assets must generate

  • track a fund’s progress towards fully funded status or track the performance of assets relative to the changes in liabilities

  • characteristics of the liabilities influence the composition of the asset portfolio


20
New cards

Properties of a valid benchmark

unambiguous - securities/weights clearly identified

investable - possible to replicate and hold the benchmark to earn its return

measurable - on a reasonably frequent and timely basis

appropriate - consistent with the PMs style or area of expertise

reflective of current investment opinions - manager should be familiar with constituent securities

specified in advance - constructed prior to evaluation period

accountable - manager should accept it.

21
New cards

Absolute return benchmark

a minimum target return

not investable, do not satisfy benchmark validity criteria

22
New cards

Factor-model-based benchmarks

are ambiguous, not specified in advance, may not be investable

(e.g. many ways to get value exposure)

23
New cards

Manager universe

manager peer group

broad group of manager with similar investment disciplines

expected to beat median manager

fails all tests of benchmark validity

24
New cards

Custom-security-based

built to reflect investment discipline of a specific manager

satisfies all benchmark validity criteria

will reflect the investment process + all constraints

25
New cards

Benchmark quality

Return on portfolio (P) = benchmark (B) + active return (A)

  • benchmark = market return (M) + incremental style return (S)

There P = M + S + A


26
New cards

Difficulties in benchmarking alternative investments

  • lack of investible market indexes

  • frequent use of leverage

  • limited liquidity

  • lack of readily available market values benchmark

  • use of IRR vs. twrr


27
New cards

Benchmarking Hedge Funds

encompass a broad range of possible strategies

may have an unlimited investment universe

rf + spread is sometimes used

fund manager universe may be used as a benchmark

28
New cards

Private equity benchmarks

mostly peer group benchmarks

Limitations

  • valuation methods may differ

  • RR depends on timing of gains/losses/CFs

PME - public market equivalents

  • base on aggregated cash flow data


29
New cards

Sharpe Ratio

risk-adjusted return of an investment or portfolio

<p>risk-adjusted return of an investment or portfolio</p>
30
New cards

Treynor Ratio

risk-adjusted performance measure that evaluates how much excess return a portfolio generates for each unit of systematic (market) risk it takes on

uses systematic risk

requires an appropriate benchmark as a proxy for systematic risk

works well for diversified portfolios


<p>risk-adjusted performance measure<span> that evaluates how much excess return a portfolio generates for each unit of </span>systematic (market) risk<span> it takes on</span></p><p>uses systematic risk</p><p>requires an appropriate benchmark as a proxy for systematic risk</p><p>works well for diversified portfolios</p><p></p>
31
New cards

Information Ratio

active return per unit of active risk

<p>active return per unit of active risk</p>
32
New cards

Appraisal Ratio

measures an investment manager's stock-picking skill by comparing a fund's alpha to its unsystematic (residual) risk

<p>measures an investment manager's stock-picking skill by comparing a fund's alpha to its unsystematic (residual) risk</p>
33
New cards

Sortino Ratio

measures an investment's risk-adjusted return relative specifically to downside risk rather than total volatility


can assess performance when return distributions are not symmetrical

Therefore, better for hedge funds & commodity trading funds

<p>measures an investment's risk-adjusted return relative specifically to downside risk rather than total volatility</p><p></p><p>can assess performance when return distributions are not symmetrical</p><p>Therefore, better for hedge funds &amp; commodity trading funds</p>
34
New cards

Capture ratios

measures the PM’s participation in up and down markets

  • upside capture (UC) - when benchmark is positive

  • downside capture (DC) - when benchmark is negative


Upside/downside ration = UC/DC

  • measures the asymmetry of return

  • > 1 = positive asymmetry (convex return profile)

  • < 1 = negative asymmetry (concave return profile)


Longer-term use case, drawdowns are more critical for short-term

Uses geometric means to measure

<p>measures the PM’s participation in up and down markets</p><ul><li><p>upside capture (UC) - when benchmark is positive</p></li><li><p>downside capture (DC) - when benchmark is negative</p></li></ul><p></p><p>Upside/downside ration = UC/DC</p><ul><li><p>measures the asymmetry of return</p></li><li><p>&gt; 1 = positive asymmetry (convex return profile)</p></li><li><p>&lt; 1 = negative asymmetry (concave return profile)</p></li></ul><p></p><p>Longer-term use case, drawdowns are more critical for short-term</p><p>Uses geometric means to measure</p>
35
New cards

Drawdown

Measures how much a portfolio has fallen from its previous peak

It is a measure of downside risk, focusing on losses rather than overall volatility

Maximum drawdown

  • cumulative peak-to-trough loss during a continuous period


Drawdown duration

  • total time from first drawdown to recovery


<p>Measures how much a portfolio has fallen from its previous peak</p><p>It is a  measure of downside risk, focusing on losses rather than overall volatility</p><p>Maximum drawdown</p><ul><li><p>cumulative peak-to-trough loss during a continuous period</p></li></ul><p></p><p>Drawdown duration</p><ul><li><p>total time from first drawdown to recovery</p></li></ul><p></p>
36
New cards

Semideviation

Denominator of the sortino ratio

<p>Denominator of the sortino ratio</p>
37
New cards

Manager Universe - Investment Manager Selection

The feasible set of investment managers, screened based on:

  • suitability

  • investment style

  • active versus passive management


38
New cards

Quantitative Factors in Management Selection

Performance track record

Attribution/appraisal

39
New cards

Qualitative Factors in Management Selection

Investment Due Diligence

  • philosophy

    • investors investment beliefs and rationale for generating returns

  • process

    • how the philosophy is implemented

  • people

  • portfolio

    • are the actual holdings and risk exposure consistent with the philosophy and process

Operational Due Diligence

  • process and procedure

    • internal controls, trade settlement, valuation etc.

  • investment vehicle

    • legal structure, custody, administration, regulatory oversight

  • terms

    • fess, liquidity, redemption restrictions setc

  • monitoring


40
New cards

Type I Error - Manager Selection

H0: manager has no skill

Rejecting the null when its true

Hiring / retaining a manager who subsequently underperforms

type I errors are errors of commission = create explicit costs (vs. Type 2 = errors of omission = create opportunity costs)

psychologically, Type I errors are more painful

41
New cards

Type II Error - Manager Selection

not hiring / firing a manager who subsequently outperforms

less likely to be measured

42
New cards

Style Analysis

Managers self-reported risk exposures are compared the the results of a returns-based or holdings-based style analysis

Should be consistent over time, deviations signal issues such as style drift

43
New cards

Returns-Based Style Analysis (Quantititative)

top-down approach

Determines the managers investment style by examining historical portfolio returns.

Involves estimating the portfolio’s sensitivities to security market indexes representing a range of distinct factors

Does not require a large amount of difficult or hard to acquire data

  • comparable across managers through time

  • provides an objective check not subject to window dressing


Drawbacks:

  • portfolio being analyzed may not reflect the current or future portfolio exposures

  • illiquid securities tend to have stale prices (smoothed returns)


44
New cards

Holdings-based style analysis (Quantitative)

Bottom-up approach

classifies the actual holdings in a portfolio at a point in time (estimates current risk factors)

comparable across managers and through time

subject to window dressing

requires manager transparency


Capture rations, drawdowns

45
New cards

Investment Philosophy

The managers underlying assumptions about the factors that drive performance

  • Passive: seek risk premiums from factor exposures

  • Active: security mispricing can be both identified and exploited

    • Behavioural: created by the actions of other market participants, temporary but numerous

    • Structural: created by internal or external rules and regulations → long lived



46
New cards

Questions to ask on investment philosophy

  1. Can the manager clearly and consistently articulate their investment philosophy

  2. Are the assumptions credible and consistent

  3. How has the philosophy developed over time

  4. Are the return sources linked to credible and consistent inefficiencies


47
New cards

Investment Personnel

Does the investment team have sufficient expertise and experience

Do they have sufficient depth

what is the level of key person risk

what kinds of agreements and incentives exist to retain and attract key employees

what has been the turnover

48
New cards

Investment Making Decision Process (4 Elements)

Signal creation (idea generation)

Signal Capture (Idea Implementation)

Portfolio Construction

Monitoring the Portfolio



49
New cards

Signal Creation (Idea Generation)

Signal creation (idea generation)

  • how are investment ideas generated

  • is the information:

    • unique (informational edge)

    • timely (timing edge)

    • implemented differently


50
New cards

Signal Capture (Idea Implementation)

Transitioning the idea into an investment position

  • what is the process

  • is it repeatable


51
New cards

Portfolio Construction

How positions are implemented

How are allocations set and adjusted

  • should be consistent with the investment philosophy and process

Are allocations consistent with conviction

  • over/under-weightings

How has growth in AUM affected portfolio characteristics

Are stop-losses used to manage risk

What types of securities are used how are hedges implemented

How are long and short ideas expressed - paired or indep.

What levels of liquidity are maintained - any limitations

52
New cards

Groupthink

occurs when a team minimizes conflict and dissent in reaching and maintaining a consensus

some members may withhold information and perspectives

(go along to get along)

Mitigation: keep team size 3-5

53
New cards

Authority bias

Group defer to a group member that is a subject matter expert or in a position of authority

Mitigation: secret ballot for votes

54
New cards

Aversion to complexity

disproportionate attention is given trivial issues at the expense of important but harder to grasp topics

aka Parkinson’s Law

Mitigation: meeting agendas drafted in advance → most important topics covered first

55
New cards

Two Broad Types of Investment Vehicles

  1. Individually managed accounts

  2. Pooled or comingled vehicle


56
New cards

Individually Managed Accounts

Advantages

  • Ownership

    • not liquidity effects from other investors

  • Customization

  • Tax efficiency

  • Transparency


Disadvantages

  • Costs

  • Tracking risk

    • If SMA is customized performance will reflect investors constraint

    • Makes attribution messy

  • Investment behaviour

    • attempts to manage the account negates the benefit of hiring a manager


57
New cards

Pooled or Comingled Vehicle

Money from multiple investors is held as a single portfolio and managed without potential customization for any investor


Liquidity

  • various types of liquidity based on the investment vehicles (ETF most liquid to PE least liquid)

  • down side and upside - less liquid more often means higher return


Often have performance based fees


58
New cards

GIPS Standards

Globally accepted standards for calculating and presenting investment returns

Used for:

  • firms

  • asset owners

  • verifiers


Not an audit it says nothing about the legitimacy of the returns


59
New cards

GIPS 5 Objectives:

Promote investor interests and instill investor confidence

Ensure accurate and consistent data

Obtain worldwide acceptance of a single standard for calculating and presenting performance

Promote fair global competition among investment firms

Promote industry self-regulation on a global basis


60
New cards

GIPS Definition of Firm

an investment firm, subsidiary, or division held out to the public as a ‘distinct business entity’

Distinct business entity (needs all three)

  • unit, division, department, or office that is organizationally and functionally segregated from other units, divisions, departments, or offices

  • that retains discretion over the assets that it manages

  • that should have autonomy over the investment decision-making process


61
New cards

GIPS Reporting Frequency

Regular portfolios (time-weighted returns): monthly at month-end

Private market portfolios: quarterly

Pooled funds not in the composite: annually

62
New cards

Modified Dietz Approach

a method of calculating portfolio returns that adjusts for the size and timing of external cash flows (deposits and withdrawals)

Use when interim valuations at the time of cash flows are unavailable or costly

Less accurate than the two segment approach

<p>a method of calculating portfolio returns that adjusts for the size and timing of external cash flows (deposits and withdrawals)</p><p>Use when interim valuations at the time of cash flows are unavailable or costly</p><p>Less accurate than the two segment approach</p>
63
New cards

Fair Value - GIPS Definition

the amount at which an investment could be sold in an orderly, arm’s length transaction between willing parties

should be the objective, observable, unadjusted quoted market price for an identical investment in an active market as of the measurement date

If the above is not possible then in order the follow:

  1. quoted prices for similar investments in active markets

  2. quoted prices for identical or similar investments in markets that are not active

  3. market-based inputs, other than quoted prices, that are observable for investment

  4. subjective, unobservable inputs


64
New cards

Composite

a group of portfolios managed according to a similar investment mandate, objective, or strategy

firms must include all actual, discretionary, fee-paying segregated accounts in at least one of the firm’s composites


eg: Composite A: US large-cap equity, Composite B US small-cap equity, Composite C: US Equity

  • An account could fit in the US large-cap and US equity composite


65
New cards

GIPS Reporting Timeline

A firm new to GIPS must initially present at least 5 years of historical performance (or since inception if less than 5 years). Add one year annually until 10 years are shown, then maintain a rolling 10-year history, dropping the oldest year each year.


Ie if a 20 year fund decided to report GIPS they would backdate at least 5 years of reporting, then year by year build until they have 10 years.

66
New cards

Required Elements of GIPS

  • composite and benchmark annual returns for all years

  • # of portfolios (if > 6) in each composite at period end

  • amount of assets in the composite

  • total firm assets at the end of each period

  • a measure of internal dispersion of individual portfolio returns for each annual period for each composite (# p >= 6)

  • 3 yr. annualized ex-post s.d. of the composite and benchmark (if monthly composite returns are available)


67
New cards

Including Performance for an Acquired Co.

  1. substantially all the investment decision makers are employed by Firm A

  2. The decision-making process remains substantially intact and independent within Firm A

  3. Firm A has records which document and support the reported performance

  4. There must be no break in the track record between Firm A and Firm B


68
New cards

Verification Process

  1. learn the firms policies and procedures

  2. sampling of portfolios

  3. sufficient testing on the sample to ensure the firm satisfies certain fundamental requirements

  4. review selected portfolios - treatment of interest, dividends, taxes

  5. Test construction/maintenance of composites

  6. Review the firms GIPS Report for compliance


69
New cards

How to Define a Firm

adopt the broadest, most meaningful definition of a firm

Company ABC could be defined as Company ABC

70
New cards

When does a firm need to disclose the number of portfolios in a composite

When there are 6 or more portfolios