A Project Report On

Analytical study of Exotic Options and its relevance

Submitted to: Submitted by:

Dr. Birendra Prasad Sudipto Ghosh

FT-
08-745

Acknowledgement
1
I want to express my sincere thanks and gratitude to my real life project guide Dr.
Birendra Prasad who has been of immense support and guidance in enabling me to do this
project. His deep understanding and valuable insights have been of great help in the
successful completion of my project.

I would also like to thank all those people, without whose help and support I would not
have been able to do justice to the project.

Sudipto Ghosh

FT-08-745

CONTENTS

2
 Topic Page No.
Introduction 4
Different types of Exotic Options 4
Barrier Option 5
Barrier Option Pricing 8
Rainbow Option Pricing 10
Conclusion 20
Demonstration of Option Trading 21
Bibliography 38

INTRODUCTION

Plain vanilla products are Derivatives like European and American call and put options. These
are traded actively in market with huge liquidity. All these mentioned derivatives are traded on

3
exchanges only. The other non standard derivatives that are traded over-the-counter are termed
as exotic options. These options are not present in any portfolio in high percentage. But the
derivative dealers keep these derivatives in their portfolio as they are much more profitable that
the standard or plain vanilla products.

Exotic products are developed for many reasons as they often meet genuine hedging and
speculation need in the market. There are sometimes (i) tax (ii) accounting (iii) legality (iv)
regulatory reasons why fund managers and financial institutions find exotic products attractive.

Non standard American options: The American options when traded over the counter sometimes
have non standard features like

i) Early exercise may be restricted to certain dates. This instrument is then known as
Bermudan option.

ii) Early exercise may be allowed during only part of the life of the option. For example,
there may be an initial lock out period with no early exercise.

iii) The strike price may change during the life of the option.

Non standard American option can be valued using binomial tree. At each node, the test for early
exercise is adjusted to reflect the terms of the option.

Compound option: Compound options are options on options. There are four types of
compound options.

i) A call on a call

ii) A put on a call

iii) A call on a put

iv) A put on a put

Compound options have two strike prices and two exercise dates. For example, a put on a put.
On the first exercise date T1, the holder of the compound option is entitled to pay the first strike
price, K1, and receive a put option. The put option gives the holder the right to sell the underlying
asset for the second strike price K2, on the second exercise date T2. The compound option will be

4
exercised on the first exercise date only if the value of the option is greater than the first strike
price.

Chooser options: A chooser option has the feature that, after a specified period of time, the
holder can choose whether the option is a call or put. Suppose that the time when the choice is
made is T1. The value of the chooser option at this time is max(c,p)

Where c is the value of call and p is the value of put underlying the option.

Then by put-call parity

max(c,p)= max(c,c+Ke-r(T1-T2)-S1e-q(T2-T1))

= c+e-q(T2-T1)max(0,Ke-(r-q)(T2-T1)-S1)

Barrier option: Barrier options are options where the payoff depends on whether the underlying
asset’s price reaches a certain level during a certain period of time. The distinctive feature
attracts the derivative dealers towards them as they are less expensive as compared to the rest of
the options.

Here the concept of knock in option and knock out option works.

Barrier options are financial derivative contracts that are activated or extinguished when the
price of the underlying asset crosses a certain level. Most models for pricing barrier options
assume continuous monitoring of the barrier. However in practice most, if not all, barrier options
traded in markets are discretely monitored. There are two problems when we discuss the discrete
barrier option. First, when the barrier is discretely monitored, a numerical method may be used to
value the option. However this method will increase calculation time exponentially with the
numbers of barrier. Second, for discrete barrier option problems, the trinomial model is useful,
but it is less effective when the barrier is very close to current asset price. In order to resolve
these two problems, we propose an analytical methodology that satisfies the partial differential
equation and initial condition that characterize the discrete barrier option problem. This method
increases the calculation time linearly with the numbers of barrier. Moreover, the method is
effective no matter what the asset price is. Furthermore, we also discuss the effect upon the
choices of the numbers of partition and the integral interval.

5
In the late 1980’s and early 1990’s, exotic options became more visible and popular in the over-
the-counter (OTC) market. Their users were big corporations, financial institutions, fund
managers, private bankers, and et cetera. The most popular group of exotic options is path-
dependent options, and barrier option is among this group. Barrier options are also called trigger
options, which become a regular option or get a rebate, depending on whether the barrier is
breached during the life of the option. The barrier options are essentially conditional options,
dependent on whether some barriers or triggers are breached within the lives of the options. They
are therefore path-dependent.

Barriers may be monitored continuously or discretely. Continuous monitoring of the barrier

means that the barrier is effect at any time prior to maturity. This means that the barrier condition
applies whenever the barrier is reached during the life of the option. Discrete monitoring implies
that the barrier is in effect only at discrete times, such as daily or weekly intervals. In other
words, the barrier condition applies only when the barrier is reached at the specified times, but
not at other times. Most models for pricing barrier options assume continuous monitoring of the
barrier; under this assumption, the option can often be priced in closed form.

The price of a discrete barrier option can be expressed in “closed form” in terms of multivariate
normal probabilities. The closed-form formula with the discrete barrier option will involve an n-
variate cumulative normal distribution if the barrier is monitored n times (Heynen and Kat,
1996). Since the multivariate cumulative normal density function in the analytical solution
cannot be expressed as an elementary function, it can only be evaluated by a numerical
approximation, which becomes either computationally intensive or produce less accurate
estimations as dimension increases.

The approximation methods could be used, such as standard lattice techniques. They obtain the
correct valuation asymptotically, but for discrete barrier options, convergence may be slow and
erratic, producing great errors even with thousands of time steps and millions of node
calculations. The reason is that the payoff of a barrier option is very sensitive to the position of
the barrier in the lattice. Boyle and Lau (1994) in their study of continuous barrier options
discover that potentially large pricing errors can result from a lattice, even with a large number
of steps. Ritchken (1996) develop a trinomial method, which adapts Boyle’s (1988) trinomial
lattice by utilizing a stretch parameter. Ritchken (1996) points out that his approach is less
effective when the barrier is very close to the current asset price. Figlewski and Gao (1999)
provides a approach that greatly increases the efficiency in lattice models. However, the partial
barrier options are monitored only at some specific time period, thus, it creates additional
difficulty for modified lattice models to price partial barrier options. Therefore, building a lattice

6
that deals effectively with discrete barriers is problematic. It has paid much attention to analytic
solutions than to lattice methods.

Heynen and Kat (1994, 1996), Carr (1995), Armstrong (2001) and Hull (2002) developed the
closed-form solutions for the price of various types of barrier options. However, the convergence
of the closed-form solutions is slow, and the results tend to have a large bias when the asset price
is close to the barrier. Besides the “close-to-barrier problem”, it is known that barrier option
price with continuous monitoring can be significantly different from those with discrete
monitoring. Cheuk and Vorst (1996) show that even hourly versus continuous monitoring can
make a significant difference in option value. Chance (1994), Flesaker (1991), and Kat and
Verdonk (1995), indicate that there can be great pricing errors between discrete and continuous
barrier options, even under daily monitoring of the barrier. Broadie, Glasserman, and Kou (1997)
discover a correction procedure. Their approach produces very accurate prices, as long as the
barrier is not close to the underlying asset price. Gao (1996) proposed an “adaptive mesh”
method, which overcomes some of the problems posed by the above models. Even with this
modification, the computational time increases, as the current underlying asset price gets closer
to the barrier. Further, Wei (1998) proposes an interpolation method between the formula for a
barrier option with the highest number of monitoring points that can be handled with the analytic
formula and the continuous case. In parallel with the determination of these pricing formulas,
numerical methods have been used for pricing barrier options, especially in those cases where
analytical pricing solutions remain unavailable, such as for discrete barrier options.

The following are the two difficulties. Firstly, when the barrier is discretely monitored, a
numerical method may be used to value the option. However this method will increase
calculation time exponentially with the numbers of barrier. Secondly, pricing is less effective and
erratic when the barrier is very close to the current asset price. Aiming to resolve these two
problems, we propose an analytical methodology that satisfies the partial differential equation
and initial condition that characterize the discrete barrier option problem. This method increases
the calculation time linearly with the numbers of barrier. Moreover, the method is effective no
matter what the asset price is.

The PDE Model

7
In this section, a partial differential equation (P.D.E.) approach is proposed to ease the
computational intensity. Finally, the P.D.E. methodology is applied to evaluate the discrete
barrier options.

2.1. The Black-Scholes Partial Differential Equation

Here the Black-Scholes partial differential equation which is abstracted from Ritchken (1987),
assuming that the return on the stock follows a diffusion process as described by the following
stochastic differential equation:
dS
S = μdt +σdZ (1)

Where S is the underlying stock price, μ is the drift on the stock per unit time, σ 2 is the variance
of the return on the stock per unit time and dZ is a mean zero normal random variable with DT,
or a standard Gauss-Weiner process. This process for stock prices is also known as the
Geometric Brownian motion.

Under the Black-Scholes assumptions, two duplicate portfolios must earn the same equilibrium
rate of return. Thus, under the no arbitrage condition, the value of the discrete barrier option can
be defined by the famous Black-Scholes partial differential equation:

And the solution of this partial differential equation is the Black-Scholes formula.

Knock in option- A latent option contract that begins to function as a normal option ("knocks
in") only once a certain price level is reached before expiration.

Technically, this type of contract is not an option until a certain price is met, so if the price is
never reached it is as if the contract never existed. Knock-ins is a type of barrier option that may
be either down-and-in option or an up-and-in option.

Knock out option- An option with a built in mechanism to expire worthless should a specified
price level be exceeded.

8
Binary options: A type of option in which the payoff is structured to be either a fixed amount of
compensation if the option expires in the money or nothing at all if the option expires out of the
money. These types of options are different from plain vanilla options. Also sometimes referred
to as "all-or-nothing options" or "digital options".

For example, suppose you were interested in buying binary call options for common shares of
ABC company with a strike price of $50 per share and a specified binary payoff of $500. If the
stock is trading above $50 when the expiration date is reached, you would receive the $500
payoff for your option contract. However, if the stock is trading below $50 per share at the
expiration date, you receive nothing.

Look back options: Call or put option whose strike price is not determined until the option is
exercised. At the time of exercise, the holder can exercise the option at any underlying price that
has occurred during the option's life. In the case of a call, the buyer will choose the lowest price,
and in the case of a put, the buyer will choose the highest price. The premium on such options
tends to be high since it gives the buyer great flexibility, and the writer has to take on a lot of
risk.

Valuation formulas have been produced for floating lookbacks. The value of a floating lookback
call at time zero is

Cfl=S0e-qTN(a1)-S0e-qt σ2/2(r-q) N(-a1)-Smine-rT[N(a2)- σ2/2(r-q) eY1N(-a3)]

Shout option: An exotic option that allows the holder to lock in a defined profit while
maintaining the right to continue participating in gains without a loss of locked-in monies. Shout
options can be structured so that holders of this contract have more than one opportunity to
"shout" or lock in profits. This allows holders to continue to benefit from positive market
movements without the possibility of losing already locked-in profits.

Rainbow option: Derivatives having two or more underlying assets or factors which cannot be

interpreted as a single composite asset or factor. Margrabe options are a type of rainbow options.

9
PRICING RAINBOW OPTIONS:

In research the issue of calculating accurate uni-, bi- and trivariate normal probabilities was
tackled. This has important applications in the pricing of multi-asset options, e.g. rainbow
options. Here the Black-Scholes prices of several styles of (multi-asset) rainbow options using
change-of-numeraire machinery. Hedging issues and deviations from the Black-Scholes pricing
model are also briefly considered.

Rainbow Options refer to all options whose payoff depends on more than one underlying risky
asset; each asset is referred to as a colour of the rainbow. Examples of these include:

1) “Best of assets or cash" option, delivering the maximum of two risky assets and cash at
expiry.

2) “Call on max" option, giving the holder the right to purchase the maximum asset at the
strike price at expiry.

3) “Call on min" option, giving the holder the right to purchase the minimum asset at the
strike price at expiry.

4) “Put on max" option, giving the holder the right to sell the maximum of the risky assets at
the strike price at expiry.

5) “Put on min" option, giving the holder the right to sell the minimum of the risky assets at
the strike at expiry.

6) “Put 2 and call 1", an exchange option to put a predefined risky asset and call the other
risky asset. Thus, asset 1 is called with the `strike' being asset 2.

Thus, the payoffs at expiry for rainbow European options are:

Best of assets or cash max(S1, S2..., Sn, K)

Call on max max(max(S1, S2…, Sn) – K, 0)

Call on min max(min(S1, S2…, Sn) –K, 0)

Put on max max(K- max(S1, S2…, Sn), 0)

Put on min max(K- min(S1,S2…, Sn), 0)

Put 2 and Call 1 max(S1- S2, 0)

10
Where

Si = Spot price of asset i,

K = Strike price of the rainbow option,

σi = volatility of asset i,

qi = dividend yield of asset i,

ρij = correlation coefficient of return on assets i and j,

r = the risk-free rate (NACC),

T = the term to expiry of the rainbow option.

The system for asset dynamics will be

dS/S = (r- q) dt + A dW

Where the Brownian motions are independent. A is a square root of the covariance matrix Σ, that
is AA’=Σ. As such, A is not uniquely determined, but it would be typical to take A to be the
Choleski decomposition matrix of Σ (that is, A is lower triangular). Under such a condition, A is
uniquely determined.

Let the ith row of A be ai. We will say that ai is the volatility vector for asset Si. Note that if we
were to write things where Si had a single volatility σi, then σi2=Σn j=1 a2ij so, where
the norm is the usual Euclidean norm. Also, the correlation between the returns of S i and Sj is
given by ai aj / || ai || ||aj ||.

The theory of rainbow options starts with (Margrabe 1978) and has its most significant other
development in (Stulz 1982). (Margrabe 1978) began by evaluating the option to exchange one
asset for the other at expiry. This is justifiably one of the most famous early option pricing
papers. This is conceptually like a call on the asset we are going to receive, but where the strike
is itself stochastic, and is in fact the second asset. The payoff at expiry for this European option
is:

Max (S1 - S2, 0)

This can be valued as

11
Where,

Margrabe derived this formula by developing and then solving a Black-Scholes type differential
equation. But he also gives another argument, which he credits to Stephen Ross, which with the
hindsight of modern technology, would be considered to be the most appropriate approach to the
problem. Let asset 2 be the numeraire in the market. In other words, asset 2 forms a new
currency, and asset one costs S1/S2 in that currency. The risk free rate in this market is q2. Thus
we have the option to buy asset one for a strike of 1. This has a Black-Scholes price of

Where S1/S2 is the volatility of S1/S2.. To get from a price in the new asset 2 currency to a price in
the original economy, we multiply by S2: the `exchange rate', which gives us (2).

Suppose that X is a European{style derivative with expiry date T. Since (Harrison & Pliska 1981)
it has been known that if X can be perfectly hedged (i.e. if there is a self financing portfolio of
underlying instruments which perfectly replicates the payoff of the derivative at expiry), then the
time-t value of the derivative is given by the following risk-neutral valuation formula:

where r is the riskless rate, and the symbol denotes the expectation at time t under a risk-
neutral measure Q. A measure Q is said to be risk-neutral if all discounted asset prices St = e-rtSt
are martingales under the measure Q, i.e. if the expected value of each ¹ St at an earlier time u is
its current value Su:

12
(Here we assume for the moment that S pays no dividends.)

Now let At = ert denote the bank account. Then the above can be rewritten as

The risk-neutral dynamics of Si/j are given by

The value of what are now called two asset rainbow options. First the value of the call on the
minimum of the two assets is derived, by evaluating the (rather unpleasant) bivariate integral.
Then a min-max parity argument is invoked: having a two asset rainbow maximum call and the
corresponding two asset rainbow minimum call is just the same as having two vanilla calls on the
two assets.

Finally put-call parity results are derived, enabling evaluation of the put on the minimum and the
put on the maximum. Rather than going into any details we immediately proceed to the more
general case where we derive far more pleasant ways of immediately finding any such valuation.

Many asset rainbow options

The delta's of the option in each of the n underlying should be, and extrapolating from there to
the price. What we do is construct general Martingale-style arguments for all cases n≥2 which
are in the style of the proof first found by Margrabe and Ross.

Johnson's results are stated for any number of assets. A rainbow option with n assets will require
the n-variate cumulative normal function for application of his formulae. As n increases, so the
computational effort and execution time for having such an approximation will increase
dramatically.

Maximum payoffs. We will first price the derivative that has payoff max (S1, S2… Sn), where
the Si satisfy the usual properties. In fact, this is notationally quite cumbersome, and all the ideas
are encapsulated in any reasonably small value of n, so we choose n = 4 (as we will see later, the
fourth asset will be the strike.

13
Firstly, the value of the derivative is the sum of the value of 4 other derivatives, the ith of which
pays Si(T) if Si(T) > Sj(T) for j 6= i, and 0 otherwise. Let us value the first of these; the others
will have similar values just by cycling the coefficients.

Considering the asset that pays S1(T) if S1(T) is the largest price. Now let S1 be the numeraire
asset with associated martingale measure Q1. We see that the value of the derivative is

Where Si/j(T)= Si(T)/ Sj(T)

Let σi/j = ||ai-aj|| We know that under Qj we have

Note that, and define

Also the correlation between Si/k(T) and Sj/k(T) is

14
Hence,

Where Ω1, Ω2, Ω3 and Ω4 are 3 x 3 matrices; the simplest way to think of them is that they are
initially 4 x 4 matrices, with Ωk having σij,k in the (i; j)th position, and then the kth row and kth
column are removed.

Thus, the value of the derivative that pays off the largest asset is

Best and worst of call options: Let us start with the case where the payoff is the best of assets
or cash. The payoff at expiry is max (S1; S2; S3; K). If we consider this to be the best of four
assets, where the fourth asset satisfies S4(T) = Ke-rt and has zero volatility, then we recover the
value of this option . This fourth asset not only has no volatility but also is independent of the
other three assets.

15
Now let us consider the rainbow call on the max option.

Recall, this has payoff max (max (S1; S2; S3) – K, 0). Note that

Max (max (S1; S2; S3) - K; 0) = max(max(S1; S2; S3);K) - K

= max (S1; S2; S3, K) – K

And so,

Finally, we have the rainbow call on the min option. (Recall, this has payoff max (min (S1; S2;
S3)-K; 0).) Because of the presence of both a maximum and minimum function, new ideas are
needed. As before we first value the derivative whose payoff is max (min (S1; S2; S3); S4).

If S4 is the worst performing asset, then the payoff is the second worst performing asset. For

1 ≤ i ≤ 3 the value of this payoff can be found by using asset Si as the numeraire. For example,
the value of the derivative that pays S1, if S4 is the worst and S1 the second worst performing
asset, is

If S4 is not the worst performing asset, then the payoff is S4. Now the probability that S4 is the
worst performing asset is

16
And so the value of the derivative that pays S4, if S4 is not the worst performing asset, is

Thus, the value of the derivative whose payoff is max (min (S1; S2; S3); S4) is

Hence the derivative with payoff max (min (S1; S2; S3); K) has value

And the call on the minimum has value

Finding the value of puts

This is easy, because put-call parity takes on a particularly useful role. It is always the case that

17
 Vc(K) + Ke-rt = Vp (K) + Vc(0)

Where the parentheses denotes strike. V could be an option on the minimum, the maximum, or
indeed any ordinal of the basket. If we have a formula for Vc(K), as established in one of the
previous sections, then we can evaluate Vc(0) by taking a limit as K↓ 0, either formally (using
facts of the manner

or informally (by
forcing our code to execute with a value of K which is very close to, but not equal to, 0 - thus
avoiding division by 0 problems but implicitly implementing the above-mentioned fact). By
rearranging, we have the put value.

Pricing rainbow options in reality

The model that has been developed here lies within the classical Black-Scholes framework. As is
well known, the assumptions of that framework do not hold in reality; various stylised facts
argue against that model. For vanilla options, the model is adjusted by means of the skew - this
skew exactly ensures that the price of the option in the market is exactly captured by the model.
Models which extract information from that skew and of how that skew will evolve are of
paramount importance in modern mathematical finance.

After a moment's thought one will realise what a difficult task we are faced with in applying
these skews here. Let us start by being completely naijve: we wish to mark our rainbow option
to market by using the skew of the various underlyings. Firstly, what strike do we use for the
underlying?

How does the strike of the rainbow translate into an appropriate strike for an option on a single
underlying? Secondly, suppose we somehow resolved this problem, and for a traded option,
wished to know its implied volatility? A familiar problem arises: often the option will have two,
sometimes even three different volatilities of one of the assets which recover the price (all other
inputs being fixed). To be more mathematical, the map from volatility to price is not injective, so
the concept of implied volatility is ill defined. See figure 1

The price for a call on the minimum of two assets. S1 = 2, S2 = 1,

18
 Figure 1

To see the sensitivity to the inputs, suppose to the setup in Figure 1 we add a third asset as
elaborated in Figure 2. Of course the general level of the value of the asset changes, but so does
the entire geometry of the price surface.

Another issue is that of the assumed correlation structure: again, correlation is di±cult to
measure; if there is implied data, then it will have a strike attached. Finally, the joint normality
hypothesis of returns of prices will typically be rejected. A popular approach is to use skews
from the vanilla market to infer the marginal distribution of returns for each of the individual
assets and then `glue them together' by means of a copula function. Given a multivariate
distribution of returns, rainbow options can then be priced by Monte Carlo methods.

19
 Figure 2

The price for a call on the minimum of three assets. As above, in addition S3 = 1, σ3 = 30%
fixed, correlation structure ρ12 = -70%, ρ 13 = 30%, ρ 23 = -20%

Conclusion

Exotic options are now widely used in global financial markets such as barrier options. Their
popularity calls for the development of faster and more stable numerical methods. In general, a
closed form valuation equation exists only in European options with a continuous barrier. For
discrete barrier options, some difficulties arise in the pricing process. The majority of valuation
methods are based on a lattice or other correction methods, which are limited to handle this
feature. In this thesis, we develop an alternative evaluation model to solve the problem.

In this section, we draw some conclusions from our pricing process as follows:

• When the numbers of partition are the same, the absolute and relative pricing errors increase as
the value of M increases.

• The majority of pricing errors remarkably comes from the numbers of partition in the integral
interval.

• The absolute and relative pricing errors have similar moving paths when the value of M
increases in multiple.

20
• From the continuity test, we detect that our pricing results are quite smooth at each partition
point. It means that the integral interval we choose is appropriate. The pricing model has several
advantages over the current approaches. Firstly, it has great flexibility in every kind of the
barriers that can be handled. Secondly, it can get up to a high accuracy by using more numbers of
partitions in the integral interval. Thirdly, the methodology possesses two properties of
increasing the calculation time linearly with the numbers of barrier and the evaluation is no
difference no matter what the underlying asset price is.

Below is the demonstration of the exotic trading by Bloomberg:

Bloomberg allows clients to quickly and easily analysis, value and save Exotic Options. Offering
the advantage of drawing together all the relevant information into the function. (Share price,
dividends, interest rates, currency cross-rates etc.)

This guide should help the user through some of the Exotic Options Screens.

OVX and OVXT allows the clients to value a whole range of exotic options – this guide is
designed to briefly explain what the option is, its benefits and how to enter the option on
Bloomberg.

OVX

21
OVX allows the user to select from a range of options.

22
The program can even help the user set up the input screen by allowing the user to refine his
Option selection type 2<GO>.

Eg by typing 17 <GO> to specify a “Down and in Knock-in Barrier” option, the relevant screen
will be called up with some of the relevant flags will be pre-set for the user.

Standard Option - One may purchase a standard OTC option, because either a registered option
does not exist, or alternatively the expiry and strike needs to be tailored to the clients
specification. In this example the user wishes to value an option that expires at year end.

23
The user needs to enter the correct strike price, the expiration date and volatility value. The
function will then calculate the theoretical price. (The client can also specify option type,
exercise type, dividends etc.)

Page 2 displays a what-if graph. The option can then be saved or even sent to a college or client.
Interest rate curve default can be selected using RDFL. Dividend assumptions updated by typing
3 <GO> in function or OPDF outside the function. NOTE user can choice IBES consensus
forecasts.

Please use this standard option as a benchmark, to compare with the exotic options values.

Warrant - if a warrant (issued by the company) exits, then new shares will be created on
exercise of the warrant. The effect of dilution needs to be incorporated in the calculations.

Note the user now needs to enter a “Y” to say the warrant is dilutive and also enter the issue size
of the warrant. Note the option price is fractionally reduced. If the underlying used is the warrant
in question all this information is automatically entered.

Note Executive Options are treated in a similar manner.

24
Cross-Currency Options - these are options that are priced in a different currency to that of the
underlying asset. This may be useful if the investor wants the asset priced locally to him. There
are two possible options:

Exchange Rate Floating (Flexos) - Payment to be made in the currency of the option using the
exchange rate at exercise. The function calls the current exchange rate and uses this to value the
option in US $. Note the price once translated is the same as the standard option (i.e the option is
just quoted in another currency but is effectively the same option.) The strike is still in sterling.

Exchange Rate Fixed (Quantos) - Payment is made in currency of option translated at a fixed
exchange rate. (i.e. the specified exchange rate is fixed.) The function again calls up the
exchange rate, the relate interest rate, volatility rate, but now the user must specify the
correlation between the stock and the exchange rate. (Clearly the user can over-type any of the
above, as with any screen.)

25
Barrier Options - these options are designed to allow the investor to benefit from their
expectation of share price path movement, (e.g. the share will first go down and then “Rocket”
up; or the shares going up but never past £20).

There are several types of barriers:

26
Knockout - if the share price exceeds the barrier the option is “blown out”. In this example the
barrier is 2000 and if the share price exceeds this barrier the option ceases. Sometimes the option
holder gets a rebate if the barrier is past.

In this example there is no rebate for the option holder as the rebate is set to 0 (by the user). The
client must enter the strike, the type of barrier, the barrier level, whether the barrier must be past
going up or down and any rebate.

Note the option price is dramatically reduced - therefore if the clients’ view (that the share will
go up but not as high as £20) is correct the investor will make a larger gain.

Knockin options only “kick-in” if their barriers are past. Eg If the user believes the stock will go

down first and then go up he can reduce his option cost by purchasing a “Down and In” Barrier
Option - the option only kicks in if the stock first falls below £1400 (it must then go up to be In-
the-money). Note if the stock just went up without first crossing the barrier the option holder gets
nothing (unless there is a rebate).

Note, the monitoring frequency, dates can be changed, this will impact the option price, e.g if the
test to see if the barrier is exceeded is only done once every 14 days (put 14 in the monitor box)

27
then the chances of passing the test are lowered and the knock-in option will be cheaper. Double
knockouts are “knocked out” if either barrier is exceeded or Double Knock ins only “kick in” if
either barrier is exceeded.

Asian Options are also very popular - these are averaging option. Most common is to average
the share price (or Rate). The payout is determined by deducting the average from the strike.
This option is far cheaper because the volatility of an average is lower than that of the price
itself. (The user still enters the volatility of the share price as normal.) This is why they appear
attractive to the purchaser. (Note, this is balanced by the chance of a large pay out being reduced
=> even if some news pushes up the share price the effect will be “diluted” by averaging the
price with the earlier lower prices.)

The averaging feature protects the writer and holder from “last minute” sharp spikes or share
price movement, and reduces the possibility and impact of manipulation. The averaging period
can be specified by entering averaging start and end, generally this is the same as the life of the
option. The user can also specify how the averaging is done Arithmetic (most common) or

28
Geometric. He/she should specify how often the averaging will be done - Continuously once a
Day/Week/Month or customized.

For Discrete enter “D” and then enter 4 <GO>. The pop-up menu then allows the user to specify
frequency, dates and even weightings. (In practice most options will equally weighted.) Note the
weekly/monthly averaging points are worked out from the averaging end in this case end of the
month (or next day if date is a weekend).

It is also possible to average the strike price. This gives the holder the right to purchase the share
at the average price (over the averaging time period.) Set the averaging to Strike. At expiry the
holder can purchase the share at the average price over the averaging period. As the strike is set
to this value. This is an attractive feature, which again coupled with the lower option price makes
this option fairly popular.

29
Look Back Options allow the holder to look back over the Look Back period and pick the most
advantageous value. In the example shown the holder can look back over the life of the option
and chose the lowest share price as the strike. Effectively he can purchase the shares at the low.
This is a very attractive but the option price is consequently higher. The user specifies that the
strike is floating, specifies the “Look Back” period, the monitoring frequency and the fact the
option is a Look Back (and not a Ladder). There is even the capability to use a percentage of the
minimum price. By changing the Strike to Fixed the option is changed such that the holder can
chose the highest price over the Look Back period and receive the difference between this and

30
the strike. (The strike price is fixed, but the share price is floating.) Again this ability to look
back and choose the most advantageous price is attractive, but this will be reflected in cost of the
option.

Ladders are similar to Look backs but the look back effectively works in steps.

31
In the above example the strike price will be changed to 1500 if the share price falls below 1500
(during the monitor period). If the price subsequently falls below 1400 this then becomes the
new strike etc. (If the share price fell to 1401 and then rises again then the strike remains at
1500.) This option is slightly less advantageous than straight Look backs, for this reason and
consequently valued as such 301p v 358.5p.

Chooser options give the holder the right to wait and choose whether the option is a Call or Put.
The user needs to input the choice date, at this point time the option holder must decide whether
to take a Call or Put. Clearly this allows the user to win if share goes up or down and therefore is
more attractive than a straight Option and consequently priced accordingly.

32
Compound Option is an option on an option. In this case, the holder has an Call option that
expires 30th June to purchase a Call option for £3.00 which will then give the holder the right to
purchase the shares at £16.00 on 31st December. The holder pays less up-front initially making
this a highly geared instrument. However, the over-all cost (including the first strike) will be
higher.

33
Digital Options, user must specify which type - in this case “Cash or Nothing” the holder gets
the stated payoff, if the strike price is exceeded or nothing. (In this case the second strike is not
used.) “Asset or Nothing” means that the asset is handed over if strike is broken.(See help for
other descriptions.)

Two Security Options - GLXO LN <Equity> SB/ LN <Equity> OVXT

Spread Options allow the holder to specify a relative view between two assets, (remember most
Fund Managers spend their time doing this). e.g. Glaxo will outperform his/her benchmark (say
the FTSE 100), or that it will outperform their competitor, etc. In Corporate situations the option
can also help back a view (e.g. the merger will go through or not).

34
In this example the purchaser believes that Glaxo will now out-perform SmithKline Beecham.
Note the weightings for SmithKline has been changed (by the user) to give a spread of about
zero. It is on this spread that the option is based. The option price is being calculated from the
volatility of each share and the correlation. (Note, one can calculate the share volatility back
from the option price, the correlation and the other shares volatility. By paging forward the user
can see the impact of say the option price (on different dates) as say the correlation changes.

35
Basket Options allow the user to value a basket of 2 stocks, (use CIX for larger baskets). (Also
see OVB).

36
Maximum Options will allow the holder to choice the maximum pay-out between the two
assets.

In this case the pay-out will be based on the maximum of 1 Glaxo share and 2.1592 SmithKline
Beecham shares, minus the strike price (1641p). This gives the holder two shots of getting a
good pay-out. Clearly Minimum is the lowest payout of the two. (Therefore it is cheaper.) Best
Option will pay the maximum between your 2 stocks and a pre-set cash amount. Dual Binary
options will pay a pre-set cash pay-out if both stocks exceed their associated strikes, else the
holder receives nothing.

37
 Bibliography

 Futures and Options, By John C. Hull

 International Research Journal of Finance and Economics

 Bloomberg

 “Exotic Options: Boundary Analyses”, Journal of Derivatives& Hedge Funds

 www.google.com

 www.investopedia.com

 Research paper by PETER OUWEHAND AND GRAEME WEST

38