*$.’) !+- ,&)*$ ( ,&’ % ,$ “# , !

A FINITE DIFFERENCE SCHEME TO CALCULATE THE OPTION PRICES IN STOCHASTIC VOLATILITY MODELS

Sh. Zamani & B. Zargari

Sharif University of Technology , [email protected]

Abstract: In stochastic volatility models, European option prices are solutions to parabolic differential equations. In this paper we propose a finite difference scheme for solving these equations numerically. We prove the stability and convergence of this method in norm infinity. Then we use the ADL method to separate the operators, this allows us to apply Thomas algorithm to solve the corresponding linear systems in each step.

! “# $ %

.* +* !”#$ %&’$ () : 8 – 6 %26 17 .% 6 , – )/ 0 1 23 2 3$ 4 5 –
[email protected] 1 – %26 * :>/ ;* ) 1 ADI 8 ; <=* .* : ) ! + 1 9
.%26 * A$ % : ; ?B 1 C2 D :* 0 1 !>0 $ )

2 3$ 8 !? () ! > :
1268732040128

Downloaded from ijiepm.iust.ac.ir at 16:22 IRST on Saturday November 4th 2017

Downloaded from ijiepm.iust.ac.ir at 16:22 IRST on Saturday November 4th 2017

!T ?; ) ) / Y3 1 .X1 )24″ ) ) 1 – ?1 “F/” 1 < .* 2
![>M – .; =1 )24″ 1 Z>M ! 5 ?;
.4 ) t = 0 ?; )
6 * ! LK MD !>6 0 1 H = H( ST ) “#$ [>M !( t =T ) )** R $
(\ ? 2/ 1 ])26 )
. H = ( ST − K )+ , ) ) ” _O ^ 0 1 9+* +B .1 `4 9+ 1 2M 6 )6
+1 a )6 ;* () *)2 1 60 1 :)2″O E 17

2 M. Scholes & F. Black

.
9+* E F6 8″ ) 1 ‘/ ;1 ” J5K” >6 ? 2/ G$ !3 H ; @ .)O > N “LK JMD” .* ) ! J5K Q – $ P 6 K 5″ $ T )** R $ 1 ) !4 )52 E F6 ; )0 t = 0 ?; ) $ t =T ?; ) :OS – 1 * ; K 1 F6 )0 ( MV )

/ / :
// :
. [email protected] . ! “#$ [email protected]

!6 ;1 `4 t>X 8 ? ; “_V 8” ; ( & 1 “#$ %&’$ () < k-: `4 8 2 ]4 * 1 Fs 8 – 6 u . (3) ;* % %$ vG – ; 4 +2 f 1B ?)1 _O w1$ s7 ) 4 1 6 ;1 `4 .)41 M*2 ` +B ; )B * 1 _O j X* G 3 ;* % 8 [b] .* )4 *1 “#$ %&’$ () _O (”
)/ 0 1 2 3$ 4 5 – .%26 , ![b] )4 j X*
.
)2 B” ? 2/ 1 %&’$ !”#$ %&’$ () 1 o” : JM/ 1 .4 “O E M\ “#$
F6 ()4 N2$) 6 * –
“#$

dS = Stt ( μ tdt +σ tdWt ) t>0

“#$ )2 B” {(σt ), 0 ≤ ≤tT } 6 )26 H)e
e 6 4 X O 1 sMx * M\ ; “#$ %&’$ () Q2$ .)41 -: 1 KO;1 * () .4 4 (σt ) 1 J X
696087-10681

97917-10681

Yt σt =Yt ?B 6 * [q] ? () ; 0
:* A k*:2 k<@6 Q ; `X

688467-53377

dSt = St (μ Yt dt + YtdWt )0≤t≤T (r) dYt = k(η−Yt ) dt +υ Yt dWt′

Wt .) M\ 17 k η !υ * 17 )/ μ 6
])2ρ∈ −( 1 1, ) 17 :M 1 2 )2 B” Wt′
12687301243

W _V < 6 W ,W′ 6 d W ,W′ t = ρdt 2
.* W ′
O !d ?1 1 .)41 2 )2 B” (r) Y )2 B”
Y 1 C2 “#$ `X y 3 !Yt = 0
M\ 1 Y 6 4 yM* kη ` y

e
M\ Y )26 -z$ 6)

%26 o” .4
-2p .* [h] ; q `X1 y>D – JM7 ](*
!Yt <η O ; !* -: 1 KO;1 e Y
1268732040128

Downloaded from ijiepm.iust.ac.ir at 16:22 IRST on Saturday November 4th 2017

Downloaded from ijiepm.iust.ac.ir at 16:22 IRST on Saturday November 4th 2017

dBt = rBtdt 0 ≤ t ≤ T (b) dSt =μStdt +σStdWt

N4 >1 .* +1 a r 4 ) % &’$ !σ 17
c* )Mc* ? c$ c “(B )c ” ;1 6 ) ?K?B 8; c6 @Kc $ cO c1 c3 H 9+* 4 )c c -4 * ; e0 )5 ? 1 dMD2 O 6 )2″O V <=* .* 6 ;1 2 ])41 c c !)41 4 ) f 1B e” ;1 5 +B .EG ?B * )M* 8; 1 * 1 1 EG
Mc*G 1 G e (” )4 K2 bghi (* 6
.)6 j X* ) 1 l e (” * * () N4 k>1 () *6 . 4 1 ?B N ; ( ) – 11 ) 1 6 * +1 !1V$ m 1 % ?B () – ; %$ m )$ 1 – 1 .% ; =1 O*
`+ 1 2M () 1 ? $ +B -1 ; 6 ) )4 ,
.6 4 “#$ %&’$ )2 B” !`X .* “#$ )2 B” !%&’$ !”#$ %&’$ () ; !) )4 +2K >X “#$ %&’$ () ?26 $
(bggb) -* -* !(bggn) ( >
.(bggi) ? Q ; `X %&’$ 6 4 o” ? () 2 )2 B” 6 * A k*:2 k<@6 2 )2 B” -2p .)41 -: 1 KO;1 e
“O E M !%&’$ 9+* )226 )
.)4 )2 B” 1 O !)4 ?1 [r] ; bq #” 6 ?2p 6) e0 () !% N”1 % `+ !() – 9+* ; @ !O* -/ (4 ) <1 ()
. 4 1 _O () – $ _ PD
?B 6 )4 % N6 ? () 1 – ;
. % l3$ ;1 !”#$ %&’$ () !N4 k>1 () P’ 1 `4 ‘s 6 ? $ !6 ;1 . 6
, O ; ) 1 ) 1 )1 () – .
.4 % (+ 1) – <=* 4

volatility
Levy processes
Stochastic volatility
Hull & White
Stein & Stein
Heston
Cox- Ingersoll- Ross
Wiener processes
Bates
142494-1051127

)
{
(

(
)
2
2
2
2
2
2
2
u
ρ
y
u
+
ρμ
y-
k(
η

)-
y
+
t
υ
2
x
u
+
k(
η
)-

y
ρυμ
y
+
y
y
u
u
(1-
+
ρ
+
)
υ
=0
2
x
y





















)

{

(



قیمت: تومان


دیدگاهتان را بنویسید