By Implication: Peter J Ackel

Download as pdf or txt
Download as pdf or txt
You are on page 1of 6

By Implication

Peter Jackel
First version:
This version:

9th July 2006


24th November 2010

Abstract

and set the price deflater to be todays net present value


of the forward starting annuity.
Since market quoting conventions for many asset classes
are such that option prices are compared for their relative
value in terms of the root-mean-square lognormal volatility
, it is important for any derivatives library to be able
to convert actual option prices into the equivalent implied
Black volatility figure. In addition, the implied volatility function is often also used either explicitly or implicitly in exotic pricing models or analytical approximations.
Particularly in the latter application, it is often important that the implied volatility computed by the derivatives
analytics library is of high accuracy, and is robust even
for parameter combinations that may, at first sight, seem
not to be relevant. For instance, in the displaced diffusion model [Rub83] governed by the stochastic differential
equation

Probably the most frequently executed numerical task in


practical financial mathematics is the calculation of the implied volatility number consistent with a given forward,
strike, time to expiry, and observable market price for European plain vanilla call and put options. At the same
time, this task is probably also the least documented one
in applied financial mathematics. In this document, it is
explained why it is not as easy as one might think to implement an implied volatility function that is both efficient
and robust, and a possible solution to the difficulty is suggested.

Introduction

A plain vanilla call or put option price p in the BlackScholes-Merton [BS73, Mer73] framework is given by
dS = [qS + (1 q) S0 ] dW ,
(1.2)

 h
i
p = F ln(F/K)
+ 2
(1.1) the vanilla option price for strike can also be computed

using the Black formula (1.1) with adjusted input paramei


 h
ters
2
K ln(F/K)

F S0 /q
where = 1 for call options and = 1 for put options,
K + (1q)/q S0

F := Se(rd)T , S is the current spot, K is the strike, r


q T .
is a flat continuously compounded interest rate to maturity,
d isa flat continuously compounded dividend rate, =
In order to attain a strongly pronounced negative implied

T,
is the root-mean-square lognormal volatility, T
volatility skew such as it is observed in the equity and inis the time to expiry, and () is the standard cumulative
terest rate market for low strikes, q often has to take on
normal distribution function. In the Black-Scholes-Merton
values as small as 104 . Conversely, for some high strikes
framework, the quantitity represents a discount factor to
in the FX or commodity markets, q may need to exceed
time T , and in general, might be referred to as the deflater
2. All of this means that the effective standard deviation
of the option price.
number in the Black formula (1.1) can easily be in the
Black [Bla76] extended the applicability of the geomet- range [104 %, 1000%], or possibly even outside. As a conric Brownian motion framework to (what we might call to- sequence, any implied volatility solver should be able to
day) an arbitrary numeraire by ingeniously separating the produce a comparatively, i.e. relatively, accurate figure even
price deflation from the calculation of an option value rel- for parameter combinations that mean that is a very small
ative to todays value of the numeraire, thus making (1.1) or moderately large number, since it shouldnt assume that
applicable to almost all areas of financial option valuation the returned number is used straightaway: it may yet, for

theory. To value an interest rate swaption, for instance, we instance, undergo division by q T for small or large q
evaluate (1.1) with F representing the forward swap rate, and small or large T . This clearly requires any solver termination criterion to be based on relative accuracy in , not
in function value.

Global head of credit, hybrid, inflation, and commodity derivative


analytics, ABN AMRO, 250 Bishopsgate, London EC2M 4AA, UK

Implying volatility

Equation (2.6) enables us to derive the asymptotics


x

Given an option price p, the task at hand is to find the loglim b = e /2 4/ ( /2)
(2.7)

standard deviation number that makes p equal to expres 3


x/ )
lim
b
=

+
x

(
(2.8)
sion (1.1) for the forward F and strike K. Once we
know

0
x
, the implied volatility figure is given by
= / T . So
far, it all seems easy.
using the definition of the normalised intrinsic value
Using the definitions
 x

x/2
/2
:= h(x) e e
.
(2.9)
p
x := ln ( F/K )
b :=
,
(2.1)
FK
From equation (2.8) we can see what happens as volatilthe equation we need to solve for becomes
ity approaches zero (for x 6= 0): since (y) decays more
rapidly than y n for any positive integer n as y ,
x/2
x/2
b = e ([ x/ + /2]) e
([ x/ /2]) . the Black option price does not permit for any regular ex(2.2) pansion for small volatilities. The extremely flat functional
form of b for small for x 6= 0 can also be seen in figure 1,
The special but very important case F = K reduces to
and this is where the trouble starts.
b|x=0 = 1 2( /2)
(2.3)
1.4

which allows for the exact

solution1

= 2 1

1
2 (1

x = 1/2
x = 1/4
x=0
x = -1/4
x = -1/2

1.2


)

(2.4)

1
0.8

wherein is the normalised option price that is to be


matched.

x = -1

b
0.6
0.4

2.1

Limits

0.2

The normalised option price b is a positively monotic function in [0, ) with the limits

 x
x
x/2
/2
(2.5)
b < e /2
h(x) e e

0
0

0.5

1.5

2.5

3.5

Figure 1: Normalised call ( = 1) option prices given by (2.2).

Conventional wisdom has it that the best all-round


method of choice for the root-finding of smooth functions
is Newtons algorithm. Alas, this is not always so for functions that do not permit regular expansions on a point in
the domain of iteration, or on its boundary. This unpleasant feature of the Black option formula is made worse by
the fact that it is convex for low volatilities and concave for
higher volatilities. This means that, given an arbitrary initial guess, a Newton iteration may, if the initial guess is too
low and in the convex domain, be fast forwarded to very
high volatilities, and not rarely to levels where the numerical implementation of our (normalised or conventional)
Black function no longer distinguishes the result from the
limit value for high volatilities. When the latter happens,
the iteration ceases and fails. Conversely, when the arbitrary initial guess is too high and in the concave domain,
the first step may attempt to propel volatility to the negative domain. Even when this doesnt happen, the Newton
algorithm can enter a non-convergent, or near-chaotic, cycle as discussed in [PTVF92, section 9.4]. Both of these
unfortunate accidents can be prevented by the addition of
safety controls in the iteration step, e.g. [PTVF92, routine
rtsafe]. Still, even with a safety feature, the Newton
method can still take many more iterations than one would
want it to in any time-critical derivatives valuation and risk

wherein h() is the Heaviside function.


In order to understand the asymptotic behaviour of (2.2)
from a purely technical point of view, let us recall [AS84,
(26.2.12)]


(z) = h(z) (z)
1 z12 + O z14 for |z|
z
(2.6)
.
z 2
with (z) = e /2
2. Equation (2.6) highlights a
common practical issue with the cumulative normal distribution function: when its argument z is significantly positive, as is the case here for deeply in the money options2 ,
(z) becomes indistinguishable from 1, or has only very
few digits in its numerical representation that separates it
from 1. The best way to overcome this problem is to use
an implementation of (z) that is highly accurate for negative z, and to only ever use out-of-the-money options when
implying Black volatility3 .
1

assuming that one has a highly accurate inverse cumulative normal


function available, such as the one published by Peter Acklam [Ack00]
2
A one-week-to-expiry call option struck at 50% of the spot for
=
50% corresponds to z 10.
3
This is the reason why put-call parity should never be used in applications: it is a nice theoretical result but useless when you rely on it in
your option pricing analytics.

n
management system where it is invoked literally billions or
|/|
4
4.E-06
even trillions of times every day . Ideally, when the correct
implied volatility is in the convex domain of the function,
3.E-06
we would want to converge from above (in function value),
2.E-06
when it is in the concave domain, we would want to con1.E-06
verge from below, since we are then guaranteed to never 0.001%
0.E+00
5.594%
leave the respective domain. Fortunately, the normalised
52.699%
-3.0
-1.9
-0.9

0.2
Black option formula (2.2) allows for the solution for the 254.163%
1.3
x
x
2.4
point of inflection. It is at
Figure 3: Left: residual relative difference between the correct implied
volatility and the number attained after Newton-iterating with start value
p
c =
2|x| .
(2.10) c given in (2.10) until the current relative step size is less than 104 of

700

600

500

400

300

200

100

0.001%

3.842%

32.828%

147.844%

489.898%

20%

70%

18%
14%
12%

40%

10%
8%

30%

6%
4%
2%
0%

20%
10% 0.001%
3.842%
0%

32.828%

147.844%
489.898%

2.4

1.3

0.2

-0.9

-1.9

32.828%

-3.0

147.844%
489.898%

2.4

1.3

0.2

-0.9

-1.9

-3.0

8%

12%

7%
10%

6%

8%

4%

6%

2%
0%

32.828%

147.844%
489.898%

2.4

1.3

0.2

-0.9

-1.9

-3.0

-1.9

-3.0

Where to start and what to aim for

5%

using

3%

4%
0.001%
3.842%

-0.9

Solving for roots of near-flat functions can be a nightmare


to tackle. Again, though, we are lucky: the normalised
Black function is amenable to straightforward root-finding
even in its near-flat region by simply switching to solving
for the value of that makes the logarithm of a given normalised option value equal to the logarithm of (2.2) ! In
the concave domain, i.e. for > c , this transformation
is not helpful, and we stick with solving for the original
normalised option value match. This means, given a target
normalised option value , the normalised moneyness x,
and the call-put flag , we define the objective function for
our root-finding algorithm as

( 
b
ln
if < bc

f() =
(3.1)
b
else

16%

60%
50%

0.001%
3.842%

80%

0.2

the attained volatility level. Right: the number of iterations associated


with the residual relative differences on the left. The calculations for
x = 0 have been done without iteration using equation (2.4).

The temptation is now to simply start at c and Newtoniterate until we are converged. This works fine for all >
c , but fails for many < c . The reason for this failure
is the near-flat shape of the normalised Black function for
small volatilities. If you try it, you will find that the iteration
almost grinds to a halt since the update step for the next
guess converges practically as rapidly to zero as the current
guess to the correct solution (for low volatilities and x 6= 0)
as shown in figures5 2 and 3. Clearly, we would prefer to

2.4

1.3

2%

bc = bc (x, ) := b(x, c , ) .

1%

0.001%
3.842%
32.828%

(3.2)

0%

147.844%
489.898%

2.4

1.3

0.2

-0.9

-1.9

Alas, this leaves us with a new dilemma: our simplistic initial guess c is no longer such a good idea for all < bc
since, with the function value on a logarithmic scale, the
normalised Black function is no longer convex but concave, and the first step is likely to attempt overshooting
into the domain of negative . We can overcome this issue by finding a better initial guess. In order to do this,
we use a technique similar to asymptotic matching known
in some sciences. Let us recapitulate: the functional form
of the asymptotic
expansion6 for 0 given in (2.8) is

( x/) 3 x2 which, alas, is not invertible in . The term
( x/), which is what gives rise to the near-flat behaviour,
is invertible in , though! Whats more, a function of the
form c(x) ( x/) (for some c(x) depending only on x)
will, for small enough be always larger than (2.8), which
means that, when inverted, it will give rise to an estimate
for that is lower than the target root. This means, in the
areas where it matters most, namely for small and thus

-3.0

Figure 2: The absolute difference between the correct implied volatility and the number attained after n iterations when starting at c and
simply Newton-iteratingwith start value c given in (2.10) for different
n on (, x) [105 , 2 2 3] [3, 3]. Top left: n = 5, top right:
n = 50, bottom left: n = 100, bottom right: n = 150. Note that
the -axis has been scaled non-linearly to highlight the region of interest. The calculations for x = 0 have been done without iteration using
equation (2.4).

have a method that doesnt need hundreds and hundreds of


iterations to converge to an acceptable accuracy in for
some perfectly reasonable parameter values.
4

The implied volatility function is not only used for the representation of market prices but often also implicitly in exotic pricing models
or analytical approximations.
5
Note that the zero levels at the back of all shown diagrams for nonzero x and very small represent outright calculation failures since for
those parameter combinations the normalised Black function value is
smaller than the smallest representable floating point number (whence it
was rounded down to zero). In other words, those areas in the parameter
plane are not attainable in practice.

6
assuming that we are only dealing with out-of-the-money option
prices

small option prices, using an approximation for (b ) that and


 x

is too high in value, but invertible, gives us precisely what
 q 
2
we need in order to start off the Newton algorithm on a
high (x, , ) := 2 1 ex |x|
. (3.7)
2
e 2 bc
logarithmic scale (in value) as suggested in (3.1). The only
open question is how to choose c(x), but that is easily done: Starting at (3.5), we thus iterate
we set it such that the crude approximation for the asymptotics near zero match the value at c . This gives us
n+1 = n + n
(3.8)
blow := (bc ) e

|x|
12
4

( x ) + .

(3.3) with the Newton step n = (x, n , ) given by




We can use a similar approach to improve the initial guess
ln b
if < bc
b
b0
when the given option price indicates that > c , i.e.
(3.9)
(x, , ) =

b
when the given value is larger than bc . In that case, we
else
b0
can use a functional form which, for very large , resemx
.
bles (2.7). We choose e /2 c(x)2( /2) since, as we with
x 2 1 2
12 (
0
) 2( 2 )
b
=
e
2
(3.10)

can see from (2.6), for large , 2( /2) / ( /2)


which matches (2.7). Matching the functions value at c
until |/| | n+1/n 1| <  for some given relative
to determine c(x) results in the invertible function
tolerance level . We show some results in figures 5 and 6.
x

x
bhigh := e 2 e 2qbc  2 .
(3.4)

|x|
2

|/|

A deliberate side-effect of the specific choice (3.4) is that


for x 0, it converges to the exact invertible form (2.3),
which means that we no longer need to special-handle the
case of x = 0 as we did previously for the calculations
shown in figures 2 and 3. We show the approximations for
different values of x in figure 4. Inverting the approxima0.5
0.45
0.4
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0

x=-1/2

0.5

1.5

0.25
0.2
0.15

6.E-11
5.E-11

0.0001%
24.7732%
76.2393%

4.E-11
3.E-11

154.1666%

2.E-11

259.7535%

1.E-11

394.4583%

0.E+00

x=-1

0.5

1.5

0.1

0.02

-1.071428571

-3

n
10

2x2

|x|4 ln b

0.2392%

0.3585%

0.4518%

0.4837%

0.6149%

0.6149%

0.7870%

0.7519%

0.8944%

3
2
-0.000001 -0.000006
0.000007 0.000003

-0.000010

Figure 6: Left: residual relative error |/| | n+1/n 1| after it


erating (3.8) starting with 0 given
/n 1| <  with
in (3.5) until | n+1
8
6
5
 = 10 on (, x) [10 , 2 2 10 ] [105 , 105 ]. Right: the
number of iterations associated with the relative errors on the left.

tions (3.3) and (3.4) gives us the improved initial guess for
any given normalised option value :

low (x, , ) if < bc
0 (x, , ) :=
(3.5)
high (x, , ) else
with

0.2981%

0.E+00
-0.000005 -0.000010
0.000010 0.000005 0.000000

0.0001%
1.E-09
0.1254%

0.0001%
0.1535%

0.857142857

Figure 4: The invertible approximations blow and bhigh for the normalised
Black function b given by equations (3.3) and (3.4).

low (x, , ) :=

2.785714286

0
0 0.5 1 1.5 2 2.5 3 3.5 4

489.8979%

x=-4

0.06

0.05

365.1145%

|/|

b
blow
bhigh

0.08

0.04

259.7535%

2.E-09

0.12

0.1

104.1775%

173.0265%

2.5

0.14

x=-2

52.5260%

Figure 5: Left: residual relative error |/| | n+1/n 1| after

iterating (3.8) starting with 0 given


in (3.5) until | n+1/n 1| < 
8
6
with  = 10 on (, x) [10 , 2 2 3][3, 3]. Right: the number
of iterations associated with the relative errors on the left.

b
blow
bhigh

0.0001%
17.6038%

0.35

7.E-11

b
blow
bhigh

0.3

9.E-11
8.E-11

-1.7 -2.4 -3.0


2.8 2.1 1.5 0.9 0.2 -0.4 -1.1

0.5
0.45
0.4
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0

b
blow
bhigh

1.E-10

Shaving off a few iterations

We can see in figures 5 and 6 that for small x, and < bc ,


the number of iterations required for convergence increases
(3.6) somewhat. Comparing this with the top left diagram in figure 4 tells us that this is caused by the initial guess (3.5)
4

0.35

0.7

not being that good for those parameter combinations: for

0.3
0.6

small x, and < bc , the initial guess 0 is given by in

0.25
0.5
version of blow given in equation (3.3), which in turn is
0.2
0.4
not that brilliant an approximation for small x and moderx = -1/4
0.15
0.3
x = -1/16
ate < bc . However, for those parameter combinations,
0.2
the approximation bhigh seems to be a better approxima- 0.1
0.05
0.1
tion for the normalised Black function b, at least as long
0
0
as > b(x, high (x, 0, ), ) since high (x, 0, ) is the point
0
0.02
0.04
0.06
0.08
0.1
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16

below which bhigh (, , ) as a function of gives negative 1.4


values and thus is definitely inappropriate as an approxima- 1.2
2.5
tion for b. Wouldnt it be nice if we could somehow find a 1
2
x = -1
mixed form of low and high to provide a better initial guess 0.8
x = -4
1.5
than (3.5) when [0, bc (x, )] ? It turns out we can. The 0.6

1
crucial idea here is not to try to find a better invertible ap- 0.4

0.5
proximation
for
b(x,
,
)
on
the
interval

[0,

]
with
c
0.2

c = 2|x|, but to interpolate the two inversions low and 0


0
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
0
0.01
0.02
0.03
0.04
0.05
high directly! We choose a functional form for interpolation

in := /bc that transfers full weight from low at = 0 to Figure 7: The interpolated initial guess interpolated (x, , ) given by
equation (4.7) for = 1 and different x on [0, bc (x, )].
high at = 1 in a fashion that some readers may recognize
as Gamma-correction or Gamma-interpolation, namely


of three enhancements we present in this section that help
interpolated := 1 w( bc ) low + w( bc ) high (4.1) to reduce the number of iterations required for a certain relative accuracy in implied Black volatility.
with w() := min( , 1), wherein we have omitted the
The second one is aimed at reducing the number of iterdependency on (x, , ) for clarity, and with low and
ations needed for very small option values and very small
high given by equations (3.6) and (3.7), respectively. We
values of x, i.e. near the money. It is motivated by lookchoose such that the interpolation is exact when =
ing at the asymptotic form (2.8) once again: the crucial
b(x, high (x, 0, ), ) by setting:
x 2
term is e( /) . Taking the logarithm, gives it a hyperbolic
2
x
:= high (x, 0, )
(4.2) form: ( /) . At this point, we recall that most iterative
algorithms are derived such that they work most efficiently
b := b(x, , )
(4.3) when the objective function whose root is sought is well

low
:= low (x, b , )
(4.4) represented by a low order polynomial. Hyperbolic forms
can be reasonably well represented. However, a hyperbolic

high := high (x, b , )


(4.5) form that is close to 1/2 is even better approximated by a
 .  
low order polynomial if we take its reciprocal! This leads

:= ln low
(4.6) us to the objective function
ln bbc
high
low
(
1
1
As it turns out, we cannot always use the above formulae as
ln(b) ln() if < bc
f () =
(4.9)
they stand. This is because, for very extreme ratios of forb
else
7
ward and strike , it is possible that the functional form (3.3)
for blow (x, , ) is not above, but below b(x, , ), and as and the Newton iteration becomes
. That
a consequence, here, we then may have < low
n+1 = n + n
(4.10)
case, however, is benign since then low is an excellent initial guess anyway, whence we use it when that happens. In
with the Newton step n = (x, n , ) given by
summary, the initial guess can be written as


ln ln(b) b
interpolated = (1 w ) low + w high
(4.7)
if < bc
b
b0
ln()
(x,
,
)
=
.



  ln(bc /)

ln(b
/b
)

c
else
b0
w = min max low , 0 , 1
.
high
low
(4.11)
(4.8)
Finally, we pull our third and last trick which exploits the
How well this works is shown in figure 7. The improved
fact that the first and second derivative of b(x, , ) with
initial guess formula (4.7) is the first, and most important,
respect to have the comparatively simple relationship

7
Thanks to Chris Gardner for pointing this out when, for instance,
F = 1 and K = 106 .

(exact)

(exact)

low

low

high

high

interpolated

interpolated

(exact)

(exact)

low

low

high

high

interpolated

interpolated

b00
b0

x2
3

(4.12)

This means, that, whenever we have computed the value


of the iterations objective function and its Newton step,
we can with comparatively little computational effort calculate the second order derivative terms required for the
higher order iterative root finding algorithm known as Halleys method [Pic88]: no additional cumulative normals, no
further inverse cumulative normals, not even exponentials
are required! Halleys iterative method is in general given
by adjusting the Newton step by a divisor of 1 + with
= /2 f 00/f 0 . In our application to the calculation of the
(normalised) Black implied volatility, we add some restrictions by capping and flooring the involved terms in order
to avoid numerical round-off arising for extremely small
values of x and sometimes leading to negative n or excessive Halley steps:


n
n
n+1 = n + max 1+
,

(4.13)
n
2

n := max n , 2n


00 (x, ,)
n
3
n := max 2n ff 0 (x,
,

4
n ,)
f 00
f0

b00
b0

2+ln(b)
ln(b)

b0
b

|/|

3.5E-09
3.0E-09
4

2.5E-09

2.0E-09 0.0001%

0.0001%
0.1535%

1.5E-09

0.2981%
0.4518%

1.0E-09

0.6149%

5.0E-10

0.7870%

0.0E+00
0.000010 0.000005 0.000000 -0.000005 -0.000010

0.1254%

0.2392%
0.3585%

0.4837%
2

0.6149%
0.7519%
0.8944%
0.000010

0.000005

0.000000

-0.000005

-0.000010

Figure 9: Left: residual relative error |/| | n+1/n 1|


after iterating (4.10) starting with initial guess given by
(4.7) until
| n+1/n 1| <  with  = 108 on (, x) [106 , 2 2 105 ]
[105 , 105 ]. Right: the number of iterations associated with the relative errors on the left.

3. Ensure you only ever operate on out-of-the-money option prices, if necessary by subtracting
the normalised


x/2
x
intrinsic value = h(x) e e /2 from
and switching 1 2h(x).

(4.14)
(4.15)

4. Use the initial guess formula (4.7) to start the iteration.

1{<bc } .

(4.16)

5. When the given normalised option price is below


the point of inflection bc (x, ) of the normalised Black
formula, iterate to find the root of 1/ln b(x,,) 1/ln ,
else of b(x, , ) , i.e. use the objective function (4.9).

All of these three enhancements together lead to the rapid


convergence behaviour shown in figures 8 and 9.
n

|/|

6. Use Halleys method (4.13) with restricted stepsize.

1.E-10
9.E-11
8.E-11

7.E-11

References

6.E-11 0.0001%

0.0001%
24.7732%
76.2393%

5.E-11 17.6038%
52.5260%

[Ack00]

P.J. Acklam. An algorithm for computing the inverse normal


cumulative distribution function. home.online.no/
pjacklam/notes/invnorm/index.html, June
2000. University of Oslo, Statistics Division.

Figure 8: Left: residual relative error |/| | n+1/n 1|


after iterating (4.10) starting with initial guess given by (4.7)
until
| n+1/n 1| <  with  = 108 on (, x) [106 , 2 2 3]
[3, 3]. Right: the number of iterations associated with the relative errors on the left.

[AS84]

M. Abramowitz and I.A. Stegun. Pocketbook of Mathematical Functions. Harri Deutsch, 1984. ISBN 3-87144-818-4.

[Bla76]

F. Black. The pricing of commodity contracts. Journal of


Financial Economics, 3:167179, 1976.

[BS73]

F. Black and M. Scholes. The Pricing of Options and Corporate Liabilities. Journal of Political Economy, 81:637654,
1973.

[Mer73]

R. C. Merton. Theory of Rational Option Pricing. Bell


Journal of Economics and Management Science, 4:141183,
Spring 1973.

4.E-11
3.E-11

154.1666%

2.E-11

259.7535%

1.E-11
0.E+00

394.4583%
3.0

2.1

1.3

0.4 -0.4 -1.3 -2.1 -3.0

104.1775%
173.0265%
259.7535%
2

365.1145%
489.8979%
2.785714286

0.857142857

-1.071428571

-3

Conclusion

For efficient
calculation:-

and

robust

implied

Black

volatility [Pic88]

C. A. Pickover. A note on chaos and Halleys method.


Communications of the ACM, 31(11):13261329, November 1988.

1. Ensure you have a highly accurate cumulative and in- [PTVF92] W. H. Press, S. A. Teukolsky, W. T. Vetterling, and B. P.
verse cumulative normal function8 .
Flannery. Numerical Recipes in C. Cambridge University
Press, 1992. www.nrbook.com/a/bookcpdf.php.

2. Transform input price p, forward F , and strike K to [Rub83]


normalised coordinates x = ln F/K and = pF K
with being the discount factor to payment, annuity,
or whichever other numeraire is used, respectively.
8

such as the one published by Peter Acklam [Ack00]

M. Rubinstein. Displaced diffusion option pricing. Journal


of Finance, 38:213217, March 1983.

You might also like