Section 5 - R26 - Portfolio Performance Evaluation Flashcards

1
Q

Performance Measure (Concept)

A

How was the portfolio’s performance?
- Absolute performance
- In excess over a benchmark return

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Performance Attribution (Concept)

A

How was the portfolio performance achieved?
- Explain
- What portion was due to active manager decisions
- Decompose excess return and risk

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Performance Appraisal (Concept)

A

Was the performance achieved through manager skill or luck?

It assesses the quality of a portfolio performance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Question: An EFFECTIVE performance attribution must… (List)

A
  1. Account for all of the portfolio’s risk and return
  2. Reflect the investment decision-making process
  3. Quantify the active decisions of the portfolio manager
  4. Provide a complete understanding of the excess risk / return
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Performance Attribution Aspects in terms of Risk and Return (Explain)

A
  1. Return Attribution: Impact of active investment decisions on returns
  2. Risk Attribution: Analyzes the risk consequences of those decisions
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Types of Performance Attribution (List and Explain)

A
  1. Returns Based: Easy. Uses Portfolio’s Total Return (over a period) + Identifies its sources.
  2. Holdings Based: Beggining of Period Holdings. All transactions assumed to occur at end of day. Accuracy improves when data has shorter intervals.
  3. Transactions Based: Holdings + Transactions -> Accurate trade results
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

a. Returns Based Attribution (Detail)

A
  • Uses Total Portfolio Return
  • Appropriate when underlying holdings information is not available
  • Easy to implement
  • Least Accurate
  • Most vulnerable to data manipulation
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

b. Holdings Based Return (Detail)

A
  • References beginning of period portfolio
  • Fails to capture transaction data
  • Most appropriate if turnover is low
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

c. Transactions Based Return (Detail)

A
  • Uses both Holdings + Transactions info during specified period
  • Most accurate, most difficult and time consuming to implement
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Micro Attribution (Concept)

A

Drivers do retorno do manager + se estão consistentes com o processo de investimento

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Macro Attribution (Concept)

A

Asset Owner’s tactical asset allocation and manager selection

+ Effect of the manager selection and timing decisions

(Vontade do dono em fugir do plano estratégico / SAA)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

(a) Equity Attribution Measurement Approaches

A
  1. Brinson Hood Beebower (BHB):
  2. Brinson-Fachler (BF):
  • Diferença: Alocação
  • BF considers (ΔPw-Bw)*(rb-Rb)
  • Exemplo: no BF, se eu over-aloquei em algo que performou mal, mas MENOS MAL que o benchmark, meu “valor adicionado” será positivo.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Differentiate BHB e BF

A

BHB: Allocation from an absolute (+) or (-) perspective

BF: Allocation from a relative perspective versus a total benchmark return

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

(b) Factor-Based Models (Concept)

A
  • Decompose contributions to excess returns from factors

Carhart:
(Rp-Rf) = + b1RMRF + b2SMB + b3HML + b4WML + Erro

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

(c) Fixed Income Attribution Approaches (List)

A
  1. Exposure Decomposition: Top down. Compare various items against benchmark (duration, yield curve, sectors, active decisions).
  2. Yield Curve Decomposition: Duration based. Estimate impacts of ΔYield in Portfolio v. ΔYields impact in Benchmark

a. Calculate Total Return = % Income + % Price Change
Price Change = -ModDur*ΔYield
b. Difference = Effect of Active PM decisions
c. Yield Curve Decomposition

  1. Yield Curve Decomposition: Full Repricing
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

(c.1) Fixed Income: Exposure Decomposition (Describe)

A

Top down approach. Explain active management through a hierarchy of decisions from the top to the bottom.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

(c.2) Fixed Income: Yield Curve Decomposition (Duration Based)

A

Duration based. Top-down or bottom up.

a. Calculate Total Return = % Income + % Price Change
Price Change = -ModDur*ΔYield
b. Difference = Effect of Active PM decisions
c. Yield Curve Decomposition

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

(c.3) Fixed Income: Yield Curve Decomposition (Full Repricing)

A
  1. Reprecifies all securities given ΔYield that actually happened (instead of estimates)
  2. Measure impact of those changes in portfolio’s returns

Note: Most complex

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Risk Attribution

A

a. Absolute Mandates: identifies sources of volatility

b. Benchmark Relative: Identifies sources of tracking risk

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

Investment Decision Making Process: BOTTOM UP

A

BOTTOM UP: Contribuição Marginal, sempre

a. RELATIVE: p/ o Tracking Risk (Desvio do Benchmark)

b. ABSOLUTE: p/ o Total Risk

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

Investment Decision Making Process: TOP DOWN

A

Top Down: Attribute tracking risk to relative allocation and selection decision

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

Investment Decision Making Process: FACTORS BASED

A

Factors marginal contribution to (a) tipo de risco e (b) tipo de risco específico

RELATIVE: Tracking Risk + “ACTIVE” Specific Risk

ABSOLUTE: Total Risk + Specific Risk (não idiossincrático)

23
Q

Macro Attribution: Sponsor Level (List)

A

Decision 1: Deviations from the SAA

Decision 2: Selection of investment managers

24
Q

Tracking Risk (Concept)

A

Risco em relação ao benchmark. Relativo.

25
Q

Specific Risk (Concept)

A

Risco idiossincrático do portfolio (único)

26
Q

Active Specific Risk (Concept)

A

Subset do specific risk relativo a risco ativo de decisões do benchmark

27
Q

Benchmarks (Concept)

A

Collection of securities that represents assets available to the PM.

28
Q

Market Index Benchmark (Concept)

A

Benchmark that represents the performance of a specific security market, market segment or an asset class.

29
Q

Liability Based Index Benchmark (Concept)

A

Benchmark focused in the cashflows that an asset must generate

30
Q

Valid Benchmark Properties (List)

A
  • Unambiguous (Securities/Weights Clearly Identified)
  • Investable (Replicable)
  • Measurable (On frequent / timely basis)
  • Appropriate (Consistent with PM’s Style)
  • Reflective of Current Investment Opinions
  • Specified in Advance
  • Accountable (Manager should accept it)
31
Q

Asset Based Benchmark Types (List)

A
  • Absolute Benchmark Return
  • Broad Market Indexes
  • Style Indexes
  • Factor Based Model Benchmarks
  • Returns Based (Regressão)
  • Manager Universe (Peer)
  • Customs Security Based
32
Q

Absolute Benchmark Return (Concept)

A

Ex: 5% de retorno anual

Cons:
- Not investable

33
Q

Broad Market Indexes (Concept)

A

Ex: S&P 500

Pros: Easy recognition, understandable, widely available

Cons: Inappropriate if PM deviates

34
Q

Style Indexes (Concept)

A

Ex: Value S&P 500

Pros: Often well known, easy to understand, widely available

35
Q

Factor Based Model Benchmarks (Concept)

A
  • Ambiguous
  • Not specified in advance
  • May not be investable
36
Q

Benchmark Quality (Formula)

A

(i) P = B + A
(ii) P = M + (B-M) + A
(iii) P = M + S + A

P = Portfolio
B = Benchmark
A = Active Decisions = (P-B)
B = Market + Style (B-M)
S = Style = (B-M)

37
Q

Challenges: Benchmark Alternative Investments (List)

A
  • Lack of investable market indexes
  • Usage of Leverage
  • Limited Liquidity
  • Lack of readily available market values
38
Q

Benchmark to Hedge Funds (List)

A
  • Broad indexes not suitable
  • Rf + Spread is sometimes used
  • Fund Manager Universe may be used

Cons:
- Risk and return of benchmark not representative
- Survivorship and backfill bias
- HF performance is self reported

39
Q

Real Estate Benchmark (List)

A
  • Numerous indexes
    Cons:
  • Not representative
  • Subsets of asset class
  • Returns and values are smooth / based in appraisal data
  • Performance correlated with largest assets
  • Lag
39
Q

Private Equity Benchmark (List)

A

Mostly peer group benchmarks

Cons:
- Valuation
- Funds IRR depends a lot on timing of cashflows

40
Q

Commodity Investments Benchmark (List)

A
  • Indexes based in futures
  • Vary greatly in composition and weight
  • Actual funds use leverage
41
Q

Managed Derivatives Benchmark (List)

A
  • Specific to a single investment strategy
42
Q

Distressed Securities (List)

A
  • Illiquid
  • Hard to construct an index
43
Q

True Active Return (Formula)

A

True Active Return = P - Normal Portfolio

P = Portfolio Manager Return

44
Q

Misfit Active Return (Formula)

A

Misfit Active Return = (P - Investor Benchmark)

Investor Benchmark = Named in SAA

45
Q

Appraisal Measures (List)

A
  1. Sharpe
  2. Sortino
  3. Treynor
46
Q

Sharpe Ratio (Formula)

A

Sharpe = (Rp - Rf)/σ portfolio

47
Q

Treynor Ratio (Formula)

A

Treynor = (Rp - Rf)/β

Uses systematic risk. For well diversified portfolios.

48
Q

Information Ratio (Formula)

A

IR = (Rp - Rbenchmark) / σativo

IR = Ractive / σativo

49
Q

Appraisal Ratio (Formula)

A

AR = α / σe

α = alpha = b0 da regressão = intercepto

σe = desvio-padrão do resíduo

50
Q

Sortino Ratio (Formula)

A

Sortino = Rp-Rt / σd

Rt = Target Return
σd = Semi Deviation = Std Deviation of Downside Risk

51
Q

Sortino Ratio Usage (List)

A
  • Better for Hedge Funds and Commodity Trading
  • Usage of Semi Deviation (only downside risk)
52
Q

Capture Ratios (Concept & Formula)

A

Capture Ratio = UC / DC
If > 1 é positive / convex
If < 1 é negative / concave

UC = Rportfolio / R bench
DC = Rportfolio / R bench

sendo,
UC = Upside Capture if Rbench > 0
DC = Downside Capture if Rbench < 0

Mede assimetria da captura do upside em relação ao downside