Book Read Free

Positional Option Trading (Wiley Trading)

Page 16

by Euan Sinclair


  flows are discounted at the risk-free rate. Traders use this model now and

  interpret the dividend yield and interest rate to be (imperfectly known)

  objective variables. (Note that in practice different traders will have

  different dividend estimates and marginal rates. This could, and in rare

  cases does, allow arbitrage.)

  We turn this into a model that incorporates drift by reinterpreting the

  dividend yield as a subjective drift estimate. We use our drift estimate to

  give a subjective estimate of the underlying's forward price. This model will

  also allow arbitrage, but it will be consistent with our opinions of the real

  world.

  The prices of calls in this model are

  (7.5)

  where

  (7.6)

  (7.7)

  As with any other pricing variable or parameter, it helps to have a greek to

  measure the impact of an incorrect estimate. The partial derivative of the

  subjective option price with respect to the drift is given by

  138

  (7.8)

  Now we can derive subjective values for options and see which are most

  mispriced. In this example, we consider 1-year options on a $100 stock,

  with a volatility of 30%, a drift of 10%, and zero interest rates. The risk-

  neutral BSM prices and the subjective prices are shown in Table 7.1.

  As the drift is primarily going to be a delta effect, it should be no surprise

  that the greatest absolute difference in values is in the lower strikes. If an

  investor wants to buy a fixed number of options, these might be the best

  choice, but in this case, it is probably better to just buy the underlying

  because the discrepancy would be greatest there. The largest percentage

  edge (“bang for the buck”) is in the highest strikes, so these appear best for

  a trader who wants to invest a certain dollar amount.

  TABLE 7.1 A Comparison of Risk-Neutral and Subjective Option Prices

  Call

  BSM Subjectiv

  Difference in $

  Difference as % of

  Strike Price

  e Price

  from BSM Value

  BSM Value

  80

  22.53

  32.53

  8.99

  0.38

  85

  20.09

  28.55

  8.46

  0.42

  90

  17.01

  24.88

  7.86

  0.46

  95

  14.29

  21.52

  7.23

  0.51

  100

  11.92

  18.49

  6.57

  0.55

  105

  9.88

  15.79

  5.91

  0.60

  110

  8.14

  13.41

  5.27

  0.65

  115

  6.67

  11.32

  4.65

  0.70

  120

  5.44

  9.51

  4.07

  0.75

  However, this analysis doesn't consider the different risk characteristics of

  options with different strikes.

  Now that we have a theoretical model, we can talk about the distribution of

  subjective option returns (assuming our drift and volatility estimates are

  correct).

  Distribution of Option Returns: Summary Statistics

  (7.9)

  (7.10)

  139

  is the true probability of expiring in the money, as opposed to the

  incorrect but frequently stated number,

  , which is the risk-neutral

  probability. And the drift is a large determinant of whether an option

  expires in the money. As an example of the size of the difference, consider a

  1-year 110 strike call on a $100 stock, with volatility of 30%, a drift of 10%,

  and an interest rate of zero. The risk-neutral probability of exercise is 32%,

  whereas the subjective probability is 45%. At very short timescales the

  volatility will overwhelm the drift, but in general it is bad to assume that

  the risk-neutral probabilities are indicative of anything in the real world.

  Figures 7.1 and 7.2 show how N( d4) depends on the drift and how it varies with strike.

  FIGURE 7.1 Probability of the 3-month 150 strike call expiring in the money. Stock is $100, volatility is 30%, and rates are zero.

  FIGURE 7.2 Probability of the 3-month calls expiring in the money when the return is 20%. Stock is $100, volatility is 30%, and rates are zero.

  140

  Chance of Profit = N(d5)

  (7.11)

  Also, by assuming the underlying price is lognormally distributed, we can

  easily calculate the median intrinsic value and hence profit.

  (7.12)

  (7.13)

  Further,

  (7.14)

  (7.15)

  An example is shown in Figure 7.3.

  It is possible to calculate the moments of the options returns (Ben-Meir

  and Schiff, 2012; Boyer and Vorkink, 2014; Sinclair and Brooks, 2017), but

  the equations are complicated and give no real insight. The most important

  fact is that options have significantly positive skew. It is also important to

  note that this extreme skewness occurs even when the underlying has

  normally distributed returns. This skewness is intrinsic to options and is

  not inherited from skewness of the underlying. If the underlying has non-

  normal returns, the effects on the options will be magnified. In this case,

  there won't be analytic expressions for the option moments and

  simulations will be necessary to understand the option return moments.

  141

  FIGURE 7.3 Ninetieth percentile of the profit of the 3-month 100-strike call. Stock is $100, volatility is 30%, and rates are zero (the risk-neutral call value is $5.98).

  Strike Choice

  All of the individual risk measurements given above need to be considered.

  None is sufficient on its own to determine that a particular option is “best.”

  And this is probably a good thing, because “best” or “optimal” is only

  optimal with respect to a given criterion, and trading decisions need to be

  based on more than just one criterion.

  In the specific case of strike selection (or investment selection in general) it will be impossible to give a single optimal solution. Some problems can't

  sensibly be solved this way. Think about the question, “What is the best car

  in the world?” Here, there are many plausible definitions of best. Does best mean fastest? Most luxurious? Safest? Greenest? Cheapest to buy?

  Cheapest to run? Most reliable? Within each category it is possible to make

  valid comparisons. A Ferrari is better than a Lamborghini. A Toyota is

  better than a Geo. But the argument over whether a Ferrari is better than a

  Toyota is unresolvable, because different car users have different goals and

  preferences.

  In some other situations, it is possible to define a clear and unambiguous definition of best. Consider baseball. The goal of a baseball team is to win games. The best player is the one who most helps his team to do this. The

  individual skills of hitting, throwing, catching, and running are now seen as

  components of the ability to create wins, rather than unrelated goals in

  themselves. So here, a statistic that aggregates the component skills by

  putting them on a consis
tent scale and converting them into a measure of

  wins created is very useful. You can then meaningfully compare a power-

  hitting catcher to a fast, agile shortstop. But this introduces the danger of

  overreliance on the power of this single statistic. This number will still have

  142

  methodological issues, and there will always be sampling problems with the individual component measurements. Even the best composite statistic

  should be a starting point rather than a definitive answer.

  Trading is somewhere in between these situations. The goal of making

  money is absolute but risk is personal, both in terms of how it relates to a

  given trader's edges and abilities but also in terms of risk tolerances and

  aversions. It should be obvious that different people have different levels of

  risk aversion. This could be personal or it could be because of an external

  mandate. Anyone trading someone else's money will have to conform to

  the risk preferences of the capital provider. But it is also important to

  remember that risk depends on the skills of a specific trader. Risk is all of

  the things outside our control. So, traders with different sources of edge

  will have different remaining risk factors. If one trader has an edge in

  volatility prediction and another doesn't, volatility is an edge for the first

  and a risk to the second. This is true in most of life. For a heart surgeon,

  doing a bypass is a low-risk operation. For a random person, it would be

  murder.

  So, a composite statistic for comparing risk and reward will be useful, but

  we should also not expect too much of it.

  The most well-known of these statistics is the Sharpe ratio, the ratio of

  (excess) return to the volatility of the return. It is also well-known that the

  Sharpe ratio is not perfect. It has a large sampling error, generally

  comparable in size to the estimate. It doesn't distinguish between downside

  and upside volatility. It doesn't take higher-order moments into account at

  all. These are all problems, but the specific option-related issue is that we

  will be dealing with heavily skewed returns. (Skew can be a feature, not a

  bug. Positive skew is a good reason to include long options in a portfolio.)

  The first work to address this failing was done by Hodges (1998), who

  showed the nature of the problem with a very simple example. We have two

  probability distributions, A and B, of excess returns.

  Distribution A

  Return

  −25 −15 −5 5% 15 25 35

  %

  %

  %

  %

  %

  %

  Probabilit

  0.2 0.0

  y

  0.01 0.04 0.25 0.4

  0

  5

  4 0.01

  Distribution B

  Return

  −25 −15 −5 5% 15 25 45

  %

  %

  %

  %

  %

  %

  Probabilit

  0.2 0.0

  y

  0.01 0.04 0.25 0.4

  0

  5

  4 0.01

  Summary Statistics

  143

  Distribution

  A

  B

  Mean return

  5.00% 5.10%

  Standard

  10.00 10.34

  deviation

  %

  %

  Sharpe ratio

  0.50 0.493

  Clearly distribution B is better than distribution A. The only difference is

  that the outcome of 35% has been increased to 45%. But this (good) change

  has increased the standard deviation more than the return, so the Sharpe

  ratio of distribution B is lower than that of distribution A.

  Hodges derived a generalized Sharpe ratio (GSR) for an investor with

  exponential utility, but it was necessary to know the complete distribution

  of payoffs to make the calculation.

  Pézier (2004) applied similar reasoning to create a GSR that only requires

  the moments of the distribution (see Maillard, 2018, for a full derivation

  and discussion). His GSR is

  (7.16)

  TABLE 7.2 Projected Performance Numbers for Long Positions in Different Strike 3-Month Calls on a $100 Stock with a Drift of

  10%, Volatility of 30%, and Zero Interest Rates

  Strik Averag

  Average

  Median

  90th

  Probabilit GS

  e

  e

  Percentag Percentag Percentile y of Profit

  R

  Dollar

  e Return

  e Profit

  Percentag

  Profit

  e Profit

  80

  $2.41

  11.8%

  10.4%

  116.9%

  52%

  0.34

  85

  $2.26

  14.1%

  9.7%

  145.7%

  50%

  0.33

  90

  $2.04

  16.9%

  4.2%

  185.0%

  48%

  0.32

  95

  $1.75

  20.2%

  −13.1%

  237.6%

  44%

  0.31

  100

  $1.43

  23.8%

  −57.6%

  305.8%

  37%

  0.27

  105

  $1.10

  27.8%

  −100%

  387.8%

  31%

  0.25

  110

  $0.80

  32.0%

  −100%

  470.4%

  24%

  0.23

  115

  $0.56

  36.6%

  −100%

  509.0%

  17%

  0.20

  120

  $0.37

  41.3%

  −100%

  378.3%

  12%

  0.16

  where SR is the standard Sharpe ratio, λ3 is the skewness of returns, and λ4

  is the kurtosis. For normal returns, the GSR reduces to the Sharpe ratio.

  Positive skewness increases the GSR. Negative skewness lowers the GSR.

  Any kurtosis lowers the GSR.

  144

  Table 7.2 gives the various statistics for different 3-month call options on a

  $100 stock with a return of 10%. Both realized and implied volatilities are

  30% and rates are zero. Each trader needs to choose the strike that most

  closely matches what they are looking for.

  This analysis assumes we have paid the correct volatility level for the

  options. If we pay too much, our results look much worse. Even when

  trading purely directionally, implied volatility is very important. This is

  shown in Table 7.3 where we assume that the implied volatility was 30%

  but realized volatility was only 22% (this roughly corresponds to the typical

  variance premium). This effect needs to be considered if the implied

  volatility of the strike under consideration is very different from the ATM

  volatility.

  (Interestingly, the GSR of the 120 strike is better than that of the 105, 110,

  and 115 strikes. This is because of the extreme skew of the results.)

  TABLE 7.3 Projected Performance Numbers for Long Positions in Different Strike 3-Month Calls on a $100 Stock with a Drift of<
br />
  10%, Implied Volatility of 30%, Realized Volatility of 22%, and

  Zero Interest Rates

  Strik Averag

  Average

  Median

  90th

  Probabilit GSR

  e

  e

  Percentag Percentag Percentile y of Profit

  Dollar

  e Return

  e Profit

  Percentag

  Profit

  e Profit

  80

  $2.16

  10.6%

  10.4%

  86.5%

  54%

  0.37

  85

  $1.74

  10.8%

  9.7%

  106.8%

  53%

  0.33

  90

  $1.12

  9.3%

  4.2%

  133.4%

  50%

  0.24

  95

  $0.43 4.9%

  −13.1%

  166.3%

  43%

  0.10

  100

  −$0.17

  −2.6%

  −57.6%

  201.9%

  36%

  −0.0

  4

  105

  −$0.53

  −13.3%

  −100%

  230.6%

  26%

  −0.1

  9

  110

  −$0.65

  −26.0%

  −100%

  222.1%

  18%

  −0.21

  115

  −$0.60

  −39.8%

  −100%

  100.7%

  11%

  −0.21

  120

  −$0.47

  −53.1%

  −100%

  −100%

  6%

  −0.0

  6

  Fundamental Considerations

  So far, we have assumed our forecast was only of the mean and variance.

  Sometimes we may have a more complex view. For example, this is

  145

  common in the Eurodollar market. Traders tend to forecast in discrete

  increments; for example, a 25 bp cut has a 40% chance of occurrence,

  instead of continuous outcomes, that is, a mean return of 5%.

  In these cases, each strike should be evaluated with a different subjective

  drift parameter. Although, given the trader's bias toward a certain

  probability distribution, the analysis will probably confirm only preexisting

  opinions (opinions in, opinions out).

  Traders in most other products should be careful to ask themselves if their

  forecasts of the distribution lead to expected value. It is hard to predict

  volatility. It is harder to predict return. Predicting the full distribution is

  probably a manifestation of overconfidence.

  Conclusion

  There is no simple answer to the question, “What strike should I buy?”

 

‹ Prev