Ecient Computation of Option Price Sensitivities
Using Homogeneity and other Tricks
Oliver Rei
Weierstra-Institute for Applied Analysis and Sto chastics
Mohrenstrasse 39
10117 Berlin
GERMANY
Uwe Wystup
Commerzbank Treasury and Financial Pro ducts
Neue Mainzer Strasse 32{36
60261 Frankfurt am Main
GERMANY
http://www.mathfinance.de
Preliminary version, January 25, 1999
Abstract
No front-oce software can survive without providing derivatives of options prices
with resp ect to underlying market or mo del parameters, the so called Greeks. We
present a list of common Greeks and exploit homogeneity prop erties of nancial
markets to derive relationships b etween Greeks out of which many are mo del-
indep endent. We apply our results to Europ ean style options, rainbow options,
path-dep endent options as well as options priced in Heston's sto chastic volatility
mo del and show shortcuts to avoid exorbitant and time-consuming computations
of derivatives whicheven strong symb olic calculators fail to pro duce.
partially aliated to Delft University,by supp ort of NWO Netherlands 1
2 Rei, O. and Wystup, U.
1 Intro duction
Based on homogeneity of time and price level of a nancial pro duct we can derive
relations for the options sensitivities, the so-called \Greeks". The basic market
mo del we use is the Black-Scholes mo del with sto cks paying a continuous dividend
yield and a riskless cash b ond. This mo del supp orts the homogeneity prop erties
which are valid in general, but its structure is so simple, that we can concentrate
on the essential statements of this pap er. We will also discuss how to extend out
work to more general market mo dels.
We list the commonly used Greeks and their notations. We do not claim this list
to b e complete, b ecause one can always de ne more derivatives of the option price
function.
First of all we analyze the homogeneity of a nancial market, consisting of one sto ck
and one cash b ond. The techniques we present are easily expanded to the higher
dimensional case as we show later on.
In addition we lo ok at the Greeks of Europ ean options in the Black-Scholes mo del.
Essentially it turns out, that one only needs to knowtwo Greeks in order to calculate
all the other Greeks without di erentiating.
Another interesting example is a Europ ean derivative security dep ending on two
assets. For such rainbow options the analysis of the risk due to changing correlation
of the two assets is very imp ortant. We will showhow this risk is related to
simultaneous changes of the two underlying securities.
There are several applications of these homogeneity relations.
1. It helps saving time in computing derivatives.
2. It pro duces a robust implementation compared to Greeks via di erence quo-
tients.
3. It allows to check the quality and consistency of Greeks pro duced by nite-
di erence-, tree- or Monte Carlo metho ds.
4. It admits a computation of Greeks for Monte Carlo based values.
5. It shows relationships b etween Greeks whichwouldn't b e noticed merely by
lo oking at di erence quotients.
Option Price Sensitivities 3
1.1 Notation
S sto ck price or sto ck price pro cess
B cash b ond, usually with risk free interest rate r
r risk free interest rate
q dividend yield continuously paid
volatility of one sto ck, or volatility matrix of several sto cks
correlation in the two-asset market mo del
t date of evaluation \to day"
T date of maturity
= T t maturity of an option
x sto ck price at time t
f payo function
v x;t;::: value of an option
k strike of an option
l level of an option
v partial derivation of v with resp ect to x and analogous
x
The standard normal distribution and density functions are de ned by
2
1
1
t
2
p
1 e nt =
2
Z
x
N x = nt dt 2