Portfolio Performance Evaluation Flashcards
Characteristics of Valid Benchmark
(1) Specified in advance
(2) Appropriate
(3) Measurable
(4) Unambiguous (= clear)
(5) Reflect current manager opinion
(6) Accountable
(7) Investable
type of benchmark
+ Asset-based
+ Liability-based
type of Asset-based benchmark
(1) Absolute return benchmark
(2) Manager Universes
(3) Broad Market Indices
(4) Style Indices
(5) Factor Model Based
(6) Return-based
(7) Custom Security-based
Hedge fund benchmark
(1) Broad-market index
(2) Risk free + Spread (3-6%)
(3) Hedge fund manager universe
real estates benchmark
Real estate indexes
Private Equity benchmark
Use IRR estimate of peers / Public Market Equivalent Methodology
Commodities benchmark
Future-based Commodity Index
Managed Derivatives benchmark
Specific to single investment strategy
Distressed Securities benchmark
May use Markey Index (Barclay Distressed Securities Index)
Performance measure
(1) Sharpe Ratio
(2) Treynor Measure
(3) Information Ratio (IR)
(4) Appraisal Ratio (Treynor-Black)
(5) Sortino Ratio
(6) Upside/Downside Capture Ratios
(7) Drawdown
Performance attribution
determines the key drivers that generated the account’s performance
Performance appraisal
determines whether the performance was affected primarily
+ by investment decisions,
+ by the overall market,
+ or by chance
Return attribution
evaluates the impact of the active portfolio management decisions on the fund’s investment returns
Risk attribution
the parallel of return attribution but analyzes the impact of the portfolio manager’s active investment decisions on portfolio risk.
Micro attribution
+ analyzes at the portfolio manager’s level and
+ seeks to verify that the portfolio manager did:
* what they said they would
* understand the drivers of the portfolio’s return
Macro attribution
+ analyzes investment decisions at the fund sponsor’s level;
+ it’s commonly used with institutional investing
Returns-based attribution
regresses total portfolio returns against major risk factors (risk, size, value) to identify the active bets of the manager and their impact on active returns.
+ Advantages:
* Easy to implement
* Does not require holdings data
+ Disadvantages:
* Least accurate
* can be manipulated
Holdings-based attribution
uses beginning-of-period portfolio holdings to assess the active sector/stock selection bets of the manager and their contribution to active return
+ Advantages: more accurate than returns-based
+ Disadvantages:
* Need data on fund holdings
* Mismatch due to trading effect
Transactions-based attribution
improves upon the holdings-based attribution by including the impact of any trades executed during the evaluation period
+ Advantage: Most accurate method
+ Disadvantage:
* Most time-consuming to implement
* Highest data to requirement
* Highest complexity
allocation effect
measures the value added/subtracted through the decision to overweight/underweight a segment versus the benchmark:
+ BHB model: Ai = (wi – Wi) × Bi
+ Brinson-Fachler model: Ai = (wi – Wi)(Bi – B)
selection effect
measures the value added/subtracted through selecting investments in the portfolio different from those of the benchmark
Si = Wi × (Ri – Bi)
interaction effect
measures the impact on active return of the allocation and selection effects acting together
Ii = (wi – Wi) × (Ri – Bi)
capture ratio (CR)
= UC ratio / DC ratio
Use:
+ CR>1 => convexity => rise more, fall less => outperform
+ CR<1 => concave => rise less, fall more => underperform
UC ratio
upside capture ratio
DC ratio
downside capture ratio (DC)
absolute return benchmark
minimum target return that the manager is expected to beat
Factor-model- based benchmarks
relating a specified set of factor exposures to the returns on an account
Returns-based benchmarks
portfolio returns are related to a set of factors that explain portfolio returns
custom security-based benchmarks
are designed to reflect the manager’s security allocations and investment process
Advantages:
+ Meets required benchmark properties + benchmark validity criteria
+ Allows continual monitoring of investment processes
+ Allows fund sponsors to effectively allocate risk across investment management teams
Disadvantage:
+ It can be expensive to construct and maintain.
+ A lack of transparency by the manager (e.g., hedge funds) can make it impossible to construct such a benchmark
Eg: Quỹ cổ BĐS lấy benchmark = 30% cổ BĐS + 60% giá BĐS + 10% Giá CPI vật liệu xây dựng
Performance measurement
what was the portfolio’s performance?
Performance attribution
how was the performance achieved?
Appraisal ratio (Treynor - Black ratio / Treynor Black appraisal ratio) - Da hoi mock test
= alpha / sigma e
+ Means: sigma e: portfolio’s residual or non-systematic risk
Sortino ratio
= (Aver R - target return) / target semideviation
use:
+ calculate ratio pernalized for considers the standard deviation of the downside risk
+ penalizes a manager when portfolio return is less than the MAR (minimum acceptable return) and thus is preferred when the investor’s goal is to preserve capital
+ offers the ability to more accurately assess performance when the return distribution is not symmetrical
+ use investor-specific preferences, rather than market conditions
+ measuring option writing
Treynor ratio
= (port return - risk free) / beta
+ useful evaluate systematic, not unsystematic
excess return to style
difference between the manager’s style index (benchmark) return and the market return—S can be positive or negative
P = M + S + A
P = investment manager’s portfolio return
M = return on the market index
S = excess return to style;
A = active return
3 components of portfolio return
(1) market (M)
(2) style (S) = B-M
(3) active management (A) = P-B
P = M + S + A
3 types òf Performance-Based Fees
(1) A symmetrical structure: Computed fee = Base + Sharing of performance
(2) fully exposed to the upside: Computed fee = Higher of either Base or
Base plus sharing of positive performance
(3) not fully exposed to either the downside or the upside: Computed fee = Higher of (1) Base or (2) Base plus sharing of performance, to a limit
which structure fee align with bankruptcy risk
A symmetrical structure
manager universe
broad group of managers with similar investment disciplines. Although not a benchmark, per se, a manager universe allows investors to make comparisons with the performance of other managers following similar investment disciplines. Managers are typically expected to beat the universe’s median return. Manager universes are typically formed by asset class and the investment approach within that class
good correlation between Style return (S), Active manager return (A), and E = P - M
+ A and S: correlation not significant to 0
+ E and S: correaltaion is significatn positive
absolute return benchmark
a minimum target return that the manager is expected to beat
Eg: the Euro Interbank Offered Rate + 4%
Broad market indexes
measures of broad asset class performance
Eg:
+ JP Morgan Emerging Market Bond Index (EMBI) for emerging market bonds
+ MSCI World Index for global developed market equities
Market indexes
more narrowly defined to represent investment styles within asset classes, , resulting in style indexes
Eg: the Russell 2000 Value —- Russell 1000 Growth Indexes
Factor-model-based benchmarks
can be constructed to more closely capture the investment decision-making process
Returns-based benchmarks
are like factor-model-based benchmarks in that portfolio returns are related to a set of factors that explain portfolio return
manager universe,
manager peer group
is a broad group of managers with similar investment disciplines
with investment decision making process is Top-Down and Atribution analysis is Relative, what is appropriate risk attribution approach ?
(1) Tracking risk
(2) Selection Decisions
-> Attribution tracking risk to relative allocation and selection decisions
with investment decision making process is Top-Down/factor-based and Atribution analysis is absolute, what is appropriate risk attribution approach ?
(1) Marginal contribution
(2) Total risk
-> Factor marginal contribution to total risk and specific risk
with investment decision making process is Bottom-up and Atribution analysis is absolute, what is appropriate risk attribution approach ?
(1) Position
(2) total risk
-> Position’s marginal contribution to total risk
with investment decision making process is Bottom-up and Atribution analysis is relative, what is appropriate risk attribution approach ?
(1) Position
(2) tracking risk
-> Position’s marginal contribution to tracking risk
with investment decision making process is Factor-Based and Atribution analysis is Relative, what is appropriate risk attribution approach ?
(1) Tracking risk
Factor’s marginal contribution to tracking risk and active specific risk
drawdown duration
(1) Drawdown duration is the total time from the start of the drawdown until the cumulative drawdown recovers to zero,
(2) Drawdown duration segmented into:
+ the drawdown phase (start to trough)
+ the recovery phase (trough-to-zero cumulative return).
RMRF factor âm nói lên điều gì
investment strategy is aversed the market
the positive HML say what ?
value tilt
the positive WML say what ?
momentum tilt
VN-Index + 5% là ?
A. relative benchmark
B. Absolute benchmark
A. relative benchmark
Nếu benchmark dựa vào 1 portfolio khác thì được gọi là relative benchmark
T-Bill + 3% là ?
A. relative benchmark
B. Absolute benchmark
b. Absolute benchmark
Nếu benchmark là số cụ thể, hoặc link với 1 chỉ số khác tương đối ổn định thì được goi là absolute benchmark
