5: Hedge Funds

Question 1

Risk-Return Benefits of CTAs

Answer 1

1. Diversification
2. Performance (better than long-only commodities)
3. Access to multiple markets
4. Transparency
5. Liquidity
6. Size
7. No withholding taxes
8. Low foreign exchange risk

Question 2

Futures vs. Forwards

Answer 2

Futures

1. Require daily cash settlement
2. No net liquidating value
3. Post collateral to cover daily losses

Question 3

Advantages of systematic over discretionary strategies

Answer 3

1. No emotional bias
2. Lower key person risk
3. More scalable
4. More diversified

Question 4

NFA and CFTC

Answer 4

NFA is responsible for introducing brokers, CTAs, CPOs and FCMs. It is an independent self-regulating body.

CFTC is a federal agency

The goal of both is to protect US investors trading on US or non-US exchanges.

Question 5

Multi-currency and single-currency margining

Answer 5

If clients are moving numerous currencies in various global markets then their clearing broker (CB) is using multi-currency margining. Advantage: no transactions costs related to converting currencies. Disadvantage: exposure to change in FX gains or losses.

The CB can offer single currency margining, clients deposit one currency with the broker and CB is responsible for converting client’s money into collateral that is accepted at given exchanges.

Question 6

3 terms for futures positions

Answer 6

1. Trading level
2. Funding level
3. Notional level (1 - 2)

Question 7

Margin to Equity ratio

Answer 7

Required Margin / Equity

The higher the ratio, the higher leverage levels, the higher the fund volatility

Question 8

3 forms of market efficiency

Answer 8

1. Weak form: claims that prices incorporate historical information
2. Semi-strong: claims that prices incorporate historical and currently available public info
3. Strong-form: claims that prices incorporate public and private info

Question 9

Data mining

Answer 9

assessing numerous strategies to identify one that has worked in the past

Question 10

Data snooping

Answer 10

being influenced by past findings and not using random samples for analysis

Question 11

Over-reaction and Under-reaction mechanisms

Answer 11

Under-reaction (undershoot intrinsic value):

1. Anchoring: bias to previous value or beliefs
2. Disposition effect: closing profitable positions too quickly and holding onto losing positions for too long

Over-reaction (overshoot intrinsic value):

1. Herding: join the crowd
2. Confirmation bias: favor data that confirms a thesis
3. Representative bias: look at past prices to form beliefs

Question 12

3 technical analysis approaches for trend followers

Answer 12

1. Moving average based systems: simplest and most popular, generates long. short position based on whether current price is above/below a given moving average. Ex: dual cross-over mov. avg. system which compares ST mov avg to LT mov avg.
2. Price channels: aka trading range breakouts looks at new highs or lows (ex: 52 week highs) to find indication of a continued trend
3. RSI: aka momentum oscillator system, ex: compare today’s price to price n days ago

Question 13

Capital at Risk (CaR) ratio for managed futures

Answer 13

= Total loss if positions hit stop-loss price / NAV

this ratio represents the loss incurred if each position hits stop loss
can be used by investors who have managed account platforms b/c you need this level of transparency
CaR may underestimate the risk of loss if the stop-loss level is extremely close to the current market price, since volatility can lead the futures price to gap through the stop-loss level
CaR often overstates risk b/c it does not account for offsetting long and short positions

Question 14

Value at Risk (VaR)

Answer 14

= (alpha x SD) + average return

alpha = 1.645 for 95% confidence interval
alpha = 1.96 for 97.5% confidence interval

VaR is a measure of potential loss in an investment portfolio for a given holding period and confidence level. Most often, returns are assumed to be normally distributed (parametric approach).

Ex: A 99% 1-month VaR of $5M indicates that there is a 99% chance that the portfolio’s losses over the next month will not exceed $5M and a 1% chance that they will

Question 15

Omega Ratio (OR)

Answer 15

=average of upper partial movements / average of lower partial movements

ratio of the average realized return in excess of a target return relative to the average realized loss relative to the target return

OR < 1, the investment provides fewer opportunities to earn a return that exceeds the target level

In general, OR is lower for higher target return, higher volatility, lower skewness and higher skewness

Given a target of 0, OR for CTAs is 4 and OR for MSCI World is 2

Question 16

Facts about CTA indices

Answer 16

1. B/C CTA indices represent equally weighted portfolios of CTAs, their volatility, skewness, and kurtosis are not necessarily representative of the statistics of individual CTAs
2. Both discretionary and systematic CTA indices have zero or positive skewness, likely due to CTAs’ use of stop-loss orders to reduce downside risk
3. CTAs have little exposure to equity, interest rate, commodity, credit and volatility risk factors, this reflects diversification benefits of CTAs
4. CTAs provide downside protection during periods of market stress (both discretionary and systematic indices generated positive returns in the post-internet bubble and financial crisis)
5. Discretionary CTAs are more heterogeneous than systematic CTAs
6. Discretionary CTA indices outperform systematic CTA indices but also have a higher SD

Sidey: Systematic CTA strategies outperform Discretionary CTA strategies (?)

Question 17

Potential CTA database biases

Answer 17

1. Selection Bias b/c managers voluntarily report their performances to databases
2. Look back Bias: fund may stop reporting after period of poor performance and start reporting after a period of positive performance
3. Survivorship Bias: occurs when databases removes performance of failed funds
4. Backfill/Instant history Bias: fund reports from a date forward and then backfill returns with its past performance

Sidey: Published managed futures indices are not typically affected by look-back, survivorship or backfill bias

Question 18

3 ways to benchmark performance of managed futures

Answer 18

1. Use an index of long-only futures: not a good angle b/c futures traders are likely to be both long and short
2. Peer Groups: can be benchmarked to active and passive indices. But, the active benchmarks may be exposed to selection bias b/c of selective reporting by some CTAs. Some benchmarks may not be investible and even the investible indices may suffer from “access bias” b/c some managers may elect not to be included in the index due to restrictions
3. Passive benchmarks (PBs): passive indices have been constructed mainly for trend-following (TF) strategies (these can be active or passive). MLMI is a popular passive TF index. PBs are not the best approach for non-TF and relative value systems, due to the heterogeneity of the latter two.
   sidey: Portfolios of TF managers had significant explanatory power while those of non-TF managers had little. But, non-TF generated higher alphas than TF managers.

Question 19

MLMI vs. active CTA indices

Answer 19

1. MLMI, which is a passive TF index, is more correlated with the systematic TF CTA index than with the discretionary CTA index.
2. The beta exposure to passive TF indices is responsible for < 50% of total excess return earned by TF CTAs
3. The MLMI is not a reasonable benchmark for discretionary CTAs but is one for systematic CTAs

Question 20

CTAs, volatility and gamma

Answer 20

CTAs are unlike the derivative holders of volatility swaps or a straddle who benefit from increased volatility

CTAs are long gamma (which measures the rate of change of an option’s delta as the value of its underlying asset changes), not volatility

CTAs perform well in down markets

Most CTA strategies generate their best performance when the SP500 volatility is extremely low and the next best when SP500 volatility is extremely high

CTA best performance is not observed when changes in SP500 volatility are very large, performance tends to be the worst when there are small changes in the SP500’s volatility

One reason CTAs may appear to be long volatility is b/c reported volatilities do not take into account emerging trends. Observers are often unaware of a price breakout and the emergence of a trend. This causes a delay in detection of identifying the new trend, so, reported volatility differs significantly from true volatility

Question 21

Benefits of CTAs

Answer 21

Diversification
Downside protection in stressed markets
Zero or positive skewness

Question 22

CTAs vs stocks and bonds

Answer 22

CTAs outperformed equities and had Sharpe ratios similar to that of bonds
CTAs had low correlations with equities and bonds

Question 23

Costs associated with managing portfolio of CTAs

Answer 23

1. Foregone loss carry forward: arises from CTAs assymetric fee structure in which managers only collect performance fees when the fund’s NAV exceeds the high watermark
2. Costs of liquidation and reinvestment:
   lag 1: time to review and make a redemption decison
   lag2: several weeks betw notification and striking NAV
   lag3: time betw striking the NAV and receiving cash
   lag 4: time betw receiving 1st and last round of cash3 costs incurred from liquidating/reinvesting:
   • foregone interest on dormant cash
   • foregone excess returns on uncommitted cash
   • transaction/administrative fees for closing and opening positions

Question 24

CTA funds vs. multi-CTA funds

Answer 24

Both have low maintenance, limited liability and no control over assets. But, only multi-CTA funds provide access to diversified portfolio of CTAs

Question 25

Pros and Cons of Managed Futures accounts

Answer 25

Pros:

• investors have control on their cash and can liquidate account anytime
• investors have control of capital and can choose leverage
• transparency of positions and trades which mitigates fraud

Cons:

• many mangers do not accept managed accounts, so small pool of selection for investors
• large minimum investments
• oversight to maintain account
• accounts do not have auditor or admin, unless secured by investor

Question 26

3 types of HF replication products

Answer 26

1. Factor Based: attempts to match HF’s monthly returns
2. Payoff Distribution: attempts to replicate HF’s payoff distribution. Assumes that the returns on the clone portfolio is equal to the HF’s returns if their probability distributions are equal
3. Bottom-up or algorithmic: good for well-defined strategies with systematic trading and no manager discretion as a source of return

Question 27

3 hypothesis for the decline in alpha and increase in beta in HF products

Answer 27

1. Funds Bubble Hypothesis: entry of unskilled managers
2. Capacity Constraint Hypothesis: oversupply of investment capital will reduce per capita alpha
3. Increased allocation to active funds hypothesis: increased allocation to HF will increase correlation with traditional assets, especially in extreme cases like redemptions during stress periods

Question 28

Benefits of HF replication products

Answer 28

1. Liquidity
2. Transparency
3. Lower DD and monitoring
4. Lower fees (no carry!)
5. Easy benchmarking
6. Diversification
7. Hedging capability b/c ETFs can be shorted
8. Flexibility in terms of setting risk profiles

Still: factor based and payoff distribution models did not produce satisfactory results.

Question 29

View commonality

Answer 29

HF managers’ views are represented by their exposures, which are aggregated in an index. So, the index’s overall performance is driven by the managers’ common exposures

Question 30

Exposure Inertia

Answer 30

While individual HFs may change their exposure over time, a HF index’s overall position is more stable b/c its exposure corresponds to the aggregate of many single exposures.
Note that rapidly changing factor exposures will erode exposure inertia

Question 31

5 key steps of equity long/short strategy

Answer 31

1. Idea Generation : look for good situations to invest
2. Optimal idea expression: identify available instruments to execute idea, justifiable price ranges, possible future trends
3. Size the position: higher conviction, larger position
4. Execute the trade with minimal cost (ex: TWAP, VWAP) and determine liquidity of securities traded
5. Manage the risk by conducting scenario analysis or stress testing

Question 32

2 sources of return to long/short managers

Answer 32

1. firm based informational inefficiency

2. factor based informational inefficiency (ex: Fama French)

Question 33

4 components of HF long positions

Answer 33

1. Cap gains/loss
2. Dividends received
3. Margin interest cost of longs if leveraged
4. Interest earned on excess cash or cash posted as collateral

Question 34

4 components of HG short positions

Answer 34

1. Cap gains/loss
2. Interest earned on short sale proceeds
3. Cost of borrowing shares (part of short stock rebate)
4. Dividends paid to buyers of borrowed shares

Question 35

4 approaches of quantitative EMN strategies using technical analysis:

Answer 35

1. Pairs trading : identify 2 stocks with a high LT correlation but with a temporary deviation from their normal relationship
2. Mean reversion: assumes that returns mean revert or trend i.e. exhibit momentum
3. Co-integration: prices are tied together by a common stochastic trend. Analogy of person walking his dog on a leash
4. Statistical Arbitrage: involves advanced econometrics techniques

Question 36

Z scoring (technique used by quant hedgies)

Answer 36

standardizes a set of variables into dimensionless quantities. to cross analyze metrics such as P/E, P/TBV and P/S
Relatedly, winsorizing is a process that sets outlier z scores back to an acceptable range (what black swan?)

Question 37

10 step process for quant strategies

Answer 37

1. Gather data
2. Clean and Examine data
3. Determine the model
4. Define residuals, to mitigate noise/error
5. Generate signals to decide when the algorithm should trade, this involves converting residuals into z-scores
6. Convert the signal to positions
7. Examine realized vs. forecasted performance
8. Take the model live with real capital
9. Monitor model
10. Adjust model

Question 38

Sources of returns for quant hedgies

Answer 38

1. Latency Arbitrage: exploit time lag in price discovery
2. ETF Arbitrage: value deviation betw ETF and underlying NAV
3. Momentum factor: note that momentum combined with value strategies is more effective
4. Sharpe Ratio and Capacity: inverse relationship

Question 39

Fama French (used by quant EMN hedgies)

Answer 39

3 factors of the Fama French:

1. excess returns above the markets, typically vs SP500
2. size: long small-cap, short big-cap
3. value: long high book-to-market, short low book-to-market

These are the F-F factors and they have low correlations to each other which allows them to improve risk-adjusted returns

Question 40

Great quant meltdown of 2007

Answer 40

Summer 2007, quants had bet on the continuation of positive drift of the F-F factors and their low correlation to each other.
Housing markets started to decline, subprime HFs suffered losses. Supposedly, a large multi-strategy HF liquidated while trying to post margins on subprime losses. This escalated liquidations of crowded quant strategies to meet demands for cash. The combo of size and speed of the liquidations led to significant losses in Aug. 2007.

And and Khandani research summary: firesale liquidations combined with widespread deleveraging and temporary withdrawal of risk capital led to the quant meltdown.

Question 41

Database biases that FoFs are exempt from

Answer 41

1. No survivorship bias
2. No selection bias
3. No instant history bias

Question 42

Benefits of FoFs

Answer 42

1. Diversification across managers and styles
2. Economies of scale for smaller investors in pooling capital
3. Information advantage
4. Liquidity
5. Access to top managers
6. Bargaining power with fees
7. Educational Role
8. Leverage
9. Professional Management

Question 43

Disadvantages of FoFs

Answer 43

1. Double layer of fees
2. Performance fees on pertion of portfolio
3. Higher taxes possible due to offshore accounts in certain countries
4. Lack of transparency
5. Lack of control
6. Lack of customization
7. Exposure to other investor’s CFs

Question 44

2 ways to trade distressed securities

Answer 44

1. Loan to own securities: secured loans with an equity stake or other equity-ownership rights, but no voting control
2. Tender offers: bid by a company or third party to acquire a substantial % of a company’s debt securities. 8 suggested factors by US courts: firm terms, limited period offer, active solicitation of public shareholders, solicitation for substantial % of a class of shares, public announcement of a repurchase program, exertion of pressure upon security holders to sell stock

Question 45

2 key goals of Chapter 11

Answer 45

1. Debtor rehabilitation

2. Return maximization for debtor’s creditors

Question 46

3 risk of distressed securities

Answer 46

1. Headline/reputation risk
2. Process risk
3. Market liquidity risk

Question 47

Section 363 sale

Answer 47

Allows purchasers to acquire specific assets free of liens or liabilities. This provides a tool for distressed companies seeking to sell their assets and for buyers looking to purchase assets at potentially bargain prices

Question 48

Control vs. Non-Control investing

Answer 48

Non-control managers invest in senior levels of a distressed company’s capital structure (ex: senior secured debt), they have a low risk-reward profile given that they invest relatively late in the reorganization process or when credit risks can be minimized.

Control managers invest in relatively junior securities, are active in their investment style, making operational and managerial decisions, are longer term investors and have a high risk-reward profile. They obtain control at a discount to fair value and use such control to guide or arbitrage the reorganization process. They aim to take control either during bankruptcy or by targeting a particular class of security by identifying the “First Impaired Class”: this is the senior most tranche of claims that have the highest voting authority in bankruptcy proceedings and thus control the reorganization plan. This is related to the 1/3 blocking position in an impaired class which comes from the US bankruptcy code rule requiring 2/3 vote in favor of a plan by each of class of creditors receiving less than their claimed amount in a re-org plan.

Question 49

Hierarchy for liquidiation

Answer 49

Secured creditors
Bankruptcy admin costs
Post-petition bankruptcy expenses
Owed employee wages
Owed employee benefit plan contributions
Unsecured customer deposits
Taxes
Unfunded pension liabilities
Unsecured claims
Preferred Stockholders
Common Stockholders

Question 50

Conversion Price

Answer 50

Face Value of convertible bond / Conversion Ratio

Question 51

Conversion Premium

Answer 51

Convertible Price - Parity / Parity

Question 52

Parity

Answer 52

Stock Price x Conversion Ratio

Question 53

Delta

Answer 53

Measures the sensitivity of a convertible to the changes in the value of the underlying asset. Mathematically, it is the partial derivative of the value of the convertible bond to the stock price (dCV/dS)

The delta represents the number of shares the arbitrageur should sell short to hedge the long equity position

When the stock price is high and the convertible is deep-in-the-money, it behaves more like its underlying stock price, so the delta is close to 1. Conversely, when the stock priceis low the convertible behaves more like FI, so the delta is close to 0. ATM options tend to have deltas ~.50

Question 54

Factors associated with the price of a convertible bond

Answer 54

stock price
interest rates
volatility of underlying stock
time
credit spreads

Question 55

Gamma

Answer 55

Measures the amount by which a position’s delta changes when the stock price changes by one unit. It captures the the convexity (or nonlinearity) of the convertible bond price.

Gamma is largest for ATM convertibles and gets progressively smaller for in-and-out-of-money convertibles

When gamma is large, the portfolio needs to be frequently adjusted to maintain delta neutrality. By contrast, a small gamma indicates that delta changes slowly, so the portfolio needs less adjustment

Any change in volatility will also lead to changes in gamma

Question 56

Dodd Frank 1 (DF)

Answer 56

DF requires the SEC to oblige filers to report their short sales on Form 13F

DF replaced the private investment advisor exemption with a general requirement that an adviser to any private pool of capital must register with the SEC or a state regulator

There are a few exemptions, like for advisors who exlcusively advise VC funds or those who manage private funds with less than $150M AUM. So, DF is not for everyone

DF requires accredited investors to deduct value of their primary residence when calculation net worth

There was an exemption in place: private advisors with less than 15 clients in the past year did not have have to register with the SEC. DF eliminated this exemption

Question 57

Dodd Frank 2 (DF)

Answer 57

under DF, the following managers must register with the SEC:

• manage funds 25 - 100M and maintain principle office in a state that does not require registration of investment advisors
• manage funds 25 - 100M and maintain principle office in a state where the HF would be subject to examinations as an investment advisor by that securities commissioner
• only manage HFs > 100M AUM and maintains managed accounts
• manage HFs whose AUM > 150M and does not maintain managed accounts

Question 58

Section 13D

Answer 58

a person/entity must report to the SEC when it acquires > 5% of publicly traded equity security and this reporting must be completed within 10 days of purchase

Question 59

3 types of SEC inspections

Answer 59

1. Regular
2. Cause
3. Sweep: focuses on a # of IAs in a specific region or engaged in certain activities (ex: soft dollars)

Question 60

Role of Chief Compliance Officer (CCO)

Answer 60

1. Compliance testing/reporting
2. Review of all marketing materials and other docs
3. Record keeping
4. Annual review
5. Code of Ethics: draft, approve, distribute enforce and amend as needed

Question 61

FoFs. MSHFs and diversification

Answer 61

FoFs provide diversification benefits at the strategy level and the manager level

MSHFs only provide diversification at the strategy level

HF managers within different strategies have low correlations, this is an indication of diversification benefits. However, diversifying across managers offers significant risk-return measures. Reddy et al. research: going from a single manager to a 4-manager portfolio increased Sharp and Sortino ratios by ~90%

Question 62

Standard Portfolio Analysis of Risk (SPAN)

Answer 62

SPAN is used to determine margin requirements for futures contracts.
SPAN does not add the risk of individual positions to determine portfolio risk, but rather takes into account benefits of diversification by determining overall portfolio risk by calculating the maximum loss of a portfolio, based on which margin requirements are set.

Question 63

Recall spread

Answer 63

A recall spread is what is used to protect the credit value of a convertible asset swap against early call or conversion. A recall spread sets the price at which the credit seller is required to buy back the convertible security.