Here is your teacher waiting for Steve Wynn to come on down so I could explain index options to him. He never showed so I guess that t he will have to download this lecture and figure it out like everyone else. He was a nice host, though. And he seems very interested in math. A lot of things going on in this large curved building behind me seemed to have a connection to math. Advanced Options Trading Strategies... (part 1) options basics 104 review 2014 Gary R. Evans Read chapter 7 and review lectures 8 and 9 from Econ 104 if you don t remember this stuff.
Weights INTC 0.250 BAC 0.250 MO 0.250 JWN 0.250 Sum: 1.000 DCGR Mean Var SD INTC 0.00093 0.000166 0.01288 BAC 0.00104 0.000216 0.01469 MO 0.00086 0.000080 0.00897 JWN 0.00065 0.000118 0.01088 I want you to finish So what are we to make yours so that it gets this result... we are of this? not done. Correlation Matrix INTC BAC MO JWN INTC 0.23317 0.26090 0.24442 BAC 0.30034 0.35182 MO 0.38890 JWN Covariance Matrix INTC BAC MO JWN INTC 0.0000441 0.0000301 0.0000342 BAC 0.0000396 0.0000562 MO 0.00003800000380 JWN Can you understand why... 1. This is a good portfolio as is... we have a volatility that is lower than the volatility of our least volatile stock (0.0081 compared to 0.00897)! 00897)! 2. Why these numbers in the covariance matrix must be really low! 3. What we would do if we found a number in the correlation matrix above our threshold h ld( (say 0.4 04or 0.5). 05) Weighted portfolio variance: Weighted eg edportfolio o o volatility: oa Portfolio alpha: Annualized Portfolio alpha: 6.65674E 05 0.0081589 0.000869016 0.218123077 4. What we would do to raise our alpha! 5. What we would to to get an optimal risk/yield tradeoff for these four stocks.
The relevance of this to the Long-Term Capital story Remember that professional exotic investments by Hedge Funds will select very high alpha and very high beta investments like sovereign debt, distressed debt, exchange-rate dependent d equity investments, emerging market equities, i etc. and they know that safety is in numbers... if the investments are uncorrelated. They are often funding out of spreads from a carry-trade of come kind, borrowing at 2% and looking for alphas of 6% or 7% or higher. The borrowing gives them leverage inverse to the percentage of their equity stake (10% equity stake gives them 10 to 1 leverage). Running to a higher-yield asset at times like 2013 and 2014 involves buying, say, sovereign debt, that t has a high h probability bilit of default. You should understand by now that given any portfolio variance, leverage modifies the volatility yproportionately p L 2 (V). This means that a 5% loss becomes a 50% loss in this scenario (ask students about the math of this). Of those covariances go from near zero to one in a market panic, you are dead. [Read page 188 and 149]
Standard & Poor s Estimated Average Cumulative Default Rates for various credit ratings... showing their estimates of the estimated cumulative default rate (e.g. the estimate of a CCC/C junk bond defaulting within 2 years is above 30%, within 7 years is above 45%). Source: Standard and Poor s Global Fixed Income Research, The Time Dimension of Standard & Poor s Credit Ratings, September 22, 2101, p.3 Chart 1. Remembering actual default rate probabilities of junk-rated debt from Econ 104:
Statistical Arbitrage (stat arb) Statistical Arbitrage (called stat arb)isa"quant" a strategy used by many of the largest hedge funds, like Citadel. Generally stat arb involves building an equity portfolio of mixed long and short positions in stocks. Typically the hedge fund strives to make the portfolio "market neutral," meaning that performance does not depend upon a rising or falling market. Because the return margins on these enormous portfolios are usually ypretty thin, stat arb portfolios are usually hugely leveraged (maybe 100 to 1) by cheap borrowed money (e.g. carry trade or commercial paper). Stat Arb also uses delta hedging (theoretically) so we will come back to Stat Arb also uses delta hedging (theoretically) so we will come back to discuss this in the future.
Obviously an old example Stat Arb (cont.) Dow Jones Industrial Averages 30 components AA Alcoa HD Home Depot MO Altria AIG AIG HON Honeywell MRK Merck AXP American Express HPQ Hewlett Packard MSFT Microsoft BA Boeing IBM IBM PFE Pfizer C Citicorp INTC Intel PG Procter Gamble CAT Caterpillar JNJ Johnson & Johnson T AT&T DD Du Pont JPM JP Morgan Chase UTX United Technolgies DIS Disney KO Coca Cola VZ Verizon GE General Electric MCD McDonalds WMT Wal Mart GM General Motors MMM 3M XOM Exxon Mobile Suppose we take a market basket of stocks, like the index above (but not restricted to indexes... this is an example), and suppose we are bullish on the index, but not every stock in the index. We therefore decide to go long in most stocks but short some of them (shown in red in the example above, based upon late 2007 markets). But in what quantities? This is where stat tarb begins, with a portfolio of longs and shorts, like any hedge fund. But stat arb also aspires to be market neutral.
Stat Arb Market Neutrality Statistical Arbitrage builds an equity portfolio of longs and shorts (see previous slide) but stat arb also typically aspires to be "market neutral." This means that ideally the portfolio's performance is not improved dby either a rise or decline in the overall index (or portfolio). In the previous example, this portfolio, if correctly weighted to be "market neutral," would be insensitive to a rise or fall in the DJIA. [Your teacher once asked a stab arb desk supervisor, a Mudder, whether he thought the market was going to rise or fall in emerging months. His answer was, "Who cares?"]. How do they achieve market neutrality? On the risk side, they use some very complicated algorithms to assess portfolio risk to make the portfolio risk neutral. We will learn that approach in a few weeks when we cover advanced volatility. What therefore is the basis of profitability? Generally, you have to be more or less right about the longs and the shorts. In a rising market, the longs have to rise more than the shorts. In a falling market, the shorts have to fall more than the longs.
Stat Arb Problems?? Stat arb has been seen in recent years as the coolest of the quant strategies, but in the fall of 2007, in the midst of the sub-prime meltdown, that some stat arb hd hedge funds hdb had been losing money, something they were not supposed to do. At issue (teacher's comments): 1. They are heavily leveraged. 2. Their borrowing source is (was) a troubled market. 3. Some bullish bias may have been built into some of these funds. 4. "Quantagion?"
Financial Contagion or "Quantagion" (a dangerous new phenomenon?) [Not required, but recommended, for class]: see "What Happened to the Quants in August 2007?" by Amir E. Khandani and Andrew W. Lo, September 20, 2007 draft available on http://www2.hmc.edu/~evans/khandanilo.pdf later versions possibly available on http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1015987. What happens when large hedge funds use huge amounts of leverage and start using identical or nearly identical trading strategies? First, profit ranges must narrow, which may require even more leverage. Margin calls, credit drying up, etc., may force severe correlated liquidation.
Our recent HPQ strangle Name: Date: Gary R. Evans February 26, 2014 Strangle Implied Daily Volatility Calculator Stock Symbol: HPQ Interest rate: Stock Price: 29.640 0.010 CALL PUT Month: Mar Mar Strike: 30.00 29.00 Expiration: 3/22/14 3/22/14 Price: 1.130 1.110 Days to maturity: 24 24 DTM override: 32 32 Implied daily volatility: 0.01919 0.02146 One-day time decay: 0.020 0.021 Version 3.3 August 16, 2011 I compare this to the historical daily volatility (HDV) of HPQ for 2 years, one year, 90 day and 30 day, using a model like the one we developed in HW1. If the IDV in the model to the right is substantially higher than the HDV then the option is priced high, h and not a good candidate for any kind of long position, but may be ideal for a short position like an iron condor or writing covered calls. Calculate
Reading the Options Chain IBM's stock info Out of the Money Expiration In the Bid/Ask date Strike Prices Money same as stocks Source: Out of the Money Volume & Open Interest In the Money
Reading (blowup from previous page) You can buy the IBM November 19 195 You can buy the IBM November 19 175 Call for $1.82 (OOM), which gives you Put for $8.05 (ITM), which gives you the right to buy IBM for $195 per share the right to sell IBM for $175 per share between now and Nov 19. between now and Nov 19. Note: These examples assume purchases at Best Ask. Obviously you can submit a limit order at any price. Source: Note the big Bid/Ask the less the liquidity the bigger these spreads.
$12.00 $10.0000 $8.00 Potential Call Option Values (upon expiration) This shows only what the option will be worth if held to expiration, i given the possible prices of IBM. $6.00 This is the Nov 19 (exp) IBM OOM 195 Call, purchased at $1.82 (BA) $4.00 on Sep 29, when IBM was $177.62 (last). $2.00 $0.00 $2.00 This is a bet that the stock price will rise. $4.00 175 180 185 190 195 200 205 Profit/Loss Gross value
The all-important premium on OTM options The premium for an in-the-money option converges to zero as the option approaches expiration. The premium of an out-of-the-money option can be thought h of as simply the price of the option because the option has an intrinsic value of 0 at the moment. The premium for either is a function of 1. Time to maturity (shorter is smaller) time decay 2. The underlying stock's volatility (greater is larger) 3. The degree to which the option is in the money (more is smaller)
Issue 1: Time Decay 1.60 1.40 1.20 100 1.00 This shows the actual projected time decay of a March 75 DIA call option, purchased for $1.51, when DIA was trading at $72.15 (implied daily volatility at 0.0156), 0156) calculated l using an option calculator. This assumes no change in DIA price and no change in volatility. 0.80 0.60 0.40 0.20 0.00 33 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1
Issue 2: The Impact of Volatility Put Option Price Calculator Daily Volatility Stock symbol: TLT Put option: Oct 106 Date Today: 8/25/2010 Expiration Date: 10/16/2010 DTM: 52 Stock Price: Strike Price: 108.14 106.00 Daily Volatility: 0.0117 Interest Rate: 0.010 Time: 52 d1 Numerator: 0.02142 Duration Volatility: 0.08437 N(-d1): 0.3998 N(-d2): 0.4327 Option Price: 2.57 Option Premium: 2.57 Note: This is mapped in TLT Volatility in Finance>Volatility Calcs 3.50 300 3.00 2.50 2.00 1.50 1.00 050 0.50 0.00 Scenario: 20 days have passed, leaving 32, no change in price, so the graph above shows the sensitivity to volatility alone.
(Old slide from the glory days) Issue 3: Distance The premium on an option is determined by the three components listed on the last slide, (1) the volatility of the stock and the market in general, (2) spread from the strike price (whether h in the money or out), and, (3) especially for out-of-themoney options, the time before the option expires. It is possible to segregate these three in theory and empirically and the ability to do so is essential for advanced options trades. Advice: Design and use your own models!! Spread 10/9/08 DIA at 91.55 DIA Nov Call Ask Strike Ask Premium 75 17.10 0.55 80 12.25 0.70 90 4.45 2.90 Time 9-Oct-08 DIA at 91.55 DIA 94 Call Ask Oct 237 2.37 Nov 5.10 Dec 5.90 Mar 7.60 More distant is greater Note: Premiums were unusually high in Oct 08 because of volatility in the markets. Closer is greater
Strategy 1: Leveraging Long with a IWM Deep-in in-the the-money (DITM) rolling call Econ 104 strategies The IWM ETF tracks the Russell 2000 and trades at 10%. Buy this one Source: TD Ameritrade Option Chains, Oct. 6, 2011 The call option that you buy must have adequate open interest (at least a few hundred contracts).
Complex Strategy 2: Strangles You can buy the IBM November 19 195 You can buy the IBM November 19 175 Call for $1.82 (OOM), which gives you Put for $8.05 (ITM), which gives you the right to buy IBM for $195 per share the right to sell IBM for $175 per share between now and Nov 19. between now and Nov 19. Source: Do you remember these from the last lecture? If we do both it is a strangle!
Performance and Profitability of the Strangle $20.00 $15.00 Note: This graph is somewhat misleading. It show the profit and value of the position on the expiration date only if the option is held to expiration. It shows nothing about the possible value of an option between now and the expiration date. $10.00 $5.00 Profit Profit $0.00 $5.00 $10.00 Loss Clearly yyou are playing volatility here. This example is a little asymmetric. $15.00 150 155 160 165 170 175 180 185 190 195 200 205 210 215 220
We did this in Econ 136 in 2007. Complex Strategy 3: Goldcorp GG Hedged Covered Call (Collar) When you write an OTM call and buy an OTM put (for insurance) this is called a Collar. Goldcorp is $21.50 Nov 22.50 call option is $1.25 Nov 17.50 put option is $0.25 Buy the stock Write the call Buy the put (for insurance)
Complex Strategy 4: Writing covered calls... This above is a covered call I wrote specifically for this class last year. I bought AeroVironment (AVAV) last Spring for an Econ 136 experiment for around 26. I should have sold it when it popped above 35 but didn t. So I wrote this call for us to track until November 22. Because of high volatility this was expensive for the buyer. I pocket $1.50 per share and would actually prefer that this be exercised, allowing me to also pocket a $4 cap gain ($1.41 had I bought the stock then written the call). The $2.91 gain if executed is more than 10% absolute not bad for 50 days. If it doesn t execute I will just write another call. Note: This option was exercised.
Strategy 6: The Iron Condor cash-positive 2.00 The primary bet was This is a bet writing a strangle on low (and lower) consisting of a 132 call volatility. You are net cash 1.50 for $1.25 and a 128 put positive and want to stay for $1.64 when DIA was that way. 1.00 at $130.36. 0.50 0.00-0.50-1.00 Lower floor provided -1.50 150 by 124 Long Put, -2.00 which cost $0.80. Upper floor provided by 135 Long Call, which cost $0.35. Possible Prices of DIA on May 19, 2012-2.50 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137
Strategy 7: Butterfly spreads (call) Betting on no or little price movement: (1) Buy one ITM call, (2) buy one OTM call, (3) write two ATM calls. This is done normally for near-term expiration dates. SPY: 187.44 W: 2 187 for 2.02 B: 1 184 for 4.21 B: 1 191 for 039 0.39. Net: $1.80 Exam 2 question: When would you use this strategy and what role is played by each leg??
4.00 3.00 2.00 Butterfly (Call) payoff grid SPY Butterfly 3/4/14 SPY stock 187.44 Mar 184 Call 4.21 Buy Mar 187 Call 2.02 Write 2X Mar 191 Call 0.39 Buy Net -0.56 1.00 0.00-1.00-2.00 180 181 182 183 184 185 186 187 188 189 190 191 192 193 Net Gross
5.00 4.00 3.00 Butterfly (Put) payoff grid On a butterfly, it doesn t much matter whether you use calls or puts, payoff format is about the same. SPY Put Butterfly 3/4/14 SPY stock 187.44 Mar 184 Put 1.06 Buy Mar 187 Put 2.06 Write 2X Mar 191 Put 4.72 Buy Net -1.66 2.00 1.00 0.00-1.00-2.00 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 Net Gross
What you must do in HW4/5 STRIKE PRICE PROBABILITY CALCULATOR Analyst's s Name: Date Today: 3/14/14 Stock Stock Symbol: SPY Stock Price: 187.410 Option Option Strike Price: 190 Option Expiration Date: 3/22/14 Option Price (Best Ask): 0.630 Days to expiration: 8 Historical Data Log Growth Rate Price to Strike Price: Sample Mean Daily Growth Rate: 0.000000000000 Sample Daily Standard Deviation: Time adjusted Standard Deviation: Calculation Probability of Stock Price Greater than Strike Price at Expiration: Time is frozen to March 4, 2014, 8:52 AM. At that moment SPY is trading for 187.410. You have bought a March 22 Call option Strike Price 190 at Best Ask for $0.63. There are 8 days to expiration. You are to use the one-year historical volatility from HW1 to calculate: (1) The probability that the stock price will be higher than the strike price at expiration assuming a zero alpha, and (2) The probability that the stock price will be higher than the strike price at your estimated alpha (which may produce the same result). l)
25.00 1.20 When you ask the question, What is The NormBase file... the prob that price will go from 100 1.00 20.0000 to 105?, you are also asking How 15.00 10.0000 5.00 0.80 0.60 0.40 0.20 0.00 0.00-0.08-0.06-0.04-0.02 0 0.02 0.04 0.06 0.08 many standard deviations is 105 away from 100, and what is the probability of that? Ranges Mean to Value Cumulative Probs +/- 3 0.9973 0.4987-0.0400-2 0.0228 +/- 2 0.9545 0.4772-0.0200-1 0.1587 +/-1.5 0.8664 0.4332 0.0000 0 0.5000 +/-1 0.6827 0.3413 0.02000200 1 0.8413 +/-0.5 0.3829 0.1915 0.0600 2 0.9772 We transform 9.00 1.20 Value Cumulative Probs 90.4837-2 0.022750022750 8.00 1.00 95.1229-1 0.158655 7.00 100.0000 0 0.500000 6.00 0.80 105.1271 1 0.841345 5.00 110.5171 2 0.977250 4.00 0.60 3.00 2.00 1.00 0.40 020 0.20 Probability of X greater than: 105 is 0.164581 0.00 82 85 88 91 94 97 100 103 106 109 112 115 118 121 Helpful to understand... 0.00
Name: Date: Gary R. Evans March 4, 2014 Strangle Implied Daily Volatility Calculator Stock Symbol: HPQ Interest rate: Stock Price: 29.640 0.010 CALL PUT Month: Mar Mar Strike: 30.00 29.00 Expiration: 3/22/14 3/22/14 Price: 1.130 1.110 Days to maturity: 18 18 DTM override: 32 32... a small segment of my Visual Basic code. Where we are going... the actual strangle calculator that we used two weeks ago. Implied daily volatility: 0.01919 0.02146 One-day time decay: 0.020 0.021 Version 3.3 August 16, 2011 Calculate