My Life in a Hedge Fund

1
My Life in a Hedge Fund Marc Potters Capital
Fund Management
2
What is a Hedge Fund?

• Traditional Asset Management
• Long Only
• Equity (Stocks)
• Bonds
• Alternative Investments
• Real Estate
• Private Equity
• Hedge Funds

• Hedge Funds
• At least one of the following
• Leverage
• Short Positions
• Derivatives
• Profit from Trading

3
Hedge Fund Industry

• Number of Hedge Funds
• Total thousands
• Assets gt 1 billion 100
• Assets gt 10 billion 20
• Total Amount in Hedge Funds
• 1 1.5 trillion (1012)
• Worldwide Stock Market Capitalization
• 40 trillion

• Short Lifetime
• Typical lifetime 3y
• Survival bias
• Known Quantitative Funds
• Simmons Medallion
• Prediction Company
• Farmer and Packard
• D.E. Shaw
• Citadel

4
Types of Hedge Funds

• Long/Short Equity
• Stock Picking Short
• Event Driven
• Merger
• Distress
• Shareholder Activism
• Convertible Bonds
• Bond/Yield Arbitrage
• e.g. LTCM
• Global Macro

• Others
• Complex Derivatives
• Systematic Quantitative
• Futures Trading (CTA)
• Equity Statistical Arbitrage
• Volatility Arbitrage
• e.g. CFM

5
Short Selling

• Buy Shares Long Position
• Make money if price goes up
• Short Selling Short Position
• Borrow shares from someone else
• Sell them on market
• Need to buy them back at some future time
• Make money if price goes down

6
Futures Market

• Market for Finance Professionals
• Contract size 50-100k
• Many Asset Classes
• Stock Indices
• Government Bonds
• Short Term Interest Rates
• Currencies
• Commodities
• Oil Products
• Grains Livestock
• Metals

• Symmetric Long/Short
• Small Deposit
• 3-5 of value
• Marked-to-Market Everyday
• Anonymous
• No Counterparty Risk
• Exchange between every transaction
• Electronic Almost Everywhere

7
Call Option

• Right to buy an asset at a fixed price X (Strike)
  at a later time T (Maturity) for a payment
  upfront C (Premium).
• Premium depends on fluctuations (Volatility)
• Similar to insurance policy

• Black-Scholes Theory
• Option premium and hedge under continuous-time
  multiplicative Browninan motion
• Unrealistic assumptions
• Used as an effective theory to translate market
  prices in volatility

8
Statistical Trading

• Law of Large Numbers
• Average N
• Fluctuations vN

• Casino House
• Odd/Even 19/3751.3
• Many tables, many players, many days

• Hedge Fund
• Diversify in strategies
• Diversify in space
• Trade many uncorrelated assets (residuals)
• Avoid global factors
• Market mode
• Global volatility
• Diversify in time
• Short term trading

9
Weve got computers

• Short Term Trading Opportunities
• 100 ms
• Monitor Large Number of Assets Simultaneously
• 2000 Stocks
• 30000 Options
• Implement Sophisticated Pricing Models

• Patient
• Can wait hours to save a few dollars
• Record Everything They Do
• Audit
• Impact Analysis
• Anonymous
• No Emotion
• No Stress

10
Ingredients for a Successful Hedge Fund

• Investment Strategy
• Profit generating strategies
• Predict future prices better than chance
• Trade selection
• Risk control
• Transaction cost analysis
• Trade Execution
• Constant Innovation

• Other Departments
• IT infrastructure
• Marketing
• Back-Office
• Compliance

11
Predictable Random Walk

• How can a random walk be predictable?
• Bias (not equal to risk free rate)
• Carry trade buy currency that pays large
  interest rate
• Positive Auto-correlation
• Trend following
• Negative Auto-correlation
• Mean reverting

12
Trend Following

• Strategy
• Always Long or Short
• Go long if R(t)gtR0, go short if R(t)lt-R0,
  otherwise stay
• Make money on average if random walk is
  persistent
• Trade statistics is skewed
• Winning trades are longer than loosing trades
• Loosing trades are more frequent
• Daily return statistics is unskewed
• gauge transformation

13
Risk Control

• Fat tails
• Non-Gaussian statistics
• Leverage Effect
• Price drops induce volatility

• Volatility fluctuations
• Long range autocorrelations
• Non-trivial kurtosis
• Dynamical leverage

14
Distribution of Stock Returns
15
Risk Control

• Empirical Correlation Matrices
• N assets, T time intervals
• QN/T O(1)
• Marcenko-Pastur distribution of eigenvalues
• Small eigenvalues are dangerous for Markowitz
• w C-1 r
• Large eigenvalues may be signal
• Cut-Off cleaning
• M-P dressing of ansatz distribution

16
Empirical Eigenvalue Distribution
17
Applications of Correlation Matrices

• Generalized Pair Trading
• Pair trading compare a stock with a peer
• Create a super-peer using all correlated stocks
• Uncorrelated Residuals
• To get vN diversification need uncorrelated bets
• Invest uncorrelated portfolios with wa1
• Measure Your Risk
• Need reliable matrix out-of-sample if you
  optimize

18
Transaction Costs

• Broker Clearer Exchange fees
• Order of 0.1 to 1 bp (10-5 - 10-4)
• Bid-Ask Spread
• Order of 1 to 10 bp (10-4 - 10-3)
• Can do better with smart execution
• Price Impact
• Strongly depends on size
• Dominant cost for large funds
• Financing cost
• 0.5 to 1 per year per unit of Leverage

19
Trade Execution

• Order Book Dynamics
• Short Term Predictive Models
• Price Impact
• Market Reaction
• Experimental Physics vs Cosmology

20
Quant Crunch 1998 and 2007

• Aug. 1998
• LTCM Strategies
• Emerging Market Yield
• European Convergence
• Long Term Index Vol
• On-the-run Off-the-run
• Trigger Russian Dept Crisis

• Aug. 2007
• Quant Strategies
• High Yield Bonds (credit risk)
• Equity Quantitative
• Equity Stat Arb
• Trigger Sub-Prime Loan Crisis

21
Investment Advices

• Beware of fees!
• Dont buy complicated products
• Diversify
• Large portfolio or indices NOT a few stocks
• Include dividends in your performance analysis
• Investment managers dont beat the index
• Buy Index Funds or ETFs

• Reasonable long term targets
• RFR 3-5 is very good
• Consider impact of taxes
• Retirement accounts, life insurance etc.
• Dont invest in Hedge Funds
• Unless you know that
• They are smart
• They are not crooks
• Beware of fees!