Coupling Index and Stocks

M. Sbai
Joint work with B. Jourdain

Universite Paris-Est - CERMICS

3rd Conference on Numerical methods in Finance
Ecole des Ponts. April 15-17, 2009.

le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

1 / 32

Outline

Introduction

Model Specification

Calibration

Numerical experiments

Conclusion

le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

2 / 32

Handling both an Index and its composing stocks is still a challenging

task.
Standard approach : a model for the stocks (with smile) + a correlation
matrix (historical estimation). Then, reconstruct the index local/implied
vol. (Avellaneda et al. [1], Lee et al. [6], . . .)
Difficulty to retrieve the good shape of the index smile (steeper than stock
smile).
Dealing with the correlation matrix is tedious (keep it positive definite ?
implied correlation matrix ?).

Our objective : a new modeling approach allowing for a good fit of both Index
and stocks.
le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

3 / 32

Another viewpoint : a factor model (the index represents the market and
influences the stocks).
A new correlation structure. Correlation risk

Index vol. risk.

(In discrete time) Cizeau, Potters and Bouchaud [2001] show that it is
possible to capture the essential features of stocks cross-correlations by a
simple non-Gaussian one factor model, specially in extreme market
conditions :
ri (t) = i rm (t) + i (t)
where ri (t) =

Si (t)
Si (t1)

1 and rm is the market daily return.

Our model can be seen as an extension in continuous time.

Calibration to both index and stocks is feasible.
le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

4 / 32

Outline

Introduction

Model Specification

Calibration

Numerical experiments

Conclusion

le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

5 / 32

Consider an Index composed of M underlyings :

M

wj Stj,M

ItM =
j=1

In a risk-neutral world, we specify the following dynamics for the stocks :

j {1, . . . , M},

dStj,M
Stj,M

= (r j )dt + j (t, ItM )dBt + j (t, Stj,M )dWtj (1)

r is the short interest rate.

j [0, [ incorporates both repo cost and dividend yield of the stock j.
j is the usual beta coefficient of the stock j.
(Bt )t[0,T] , (Wt1 )t[0,T] , . . . , (WtM )t[0,T] are independent BMs.
The functions , 1 , . . . , M satisfy the usual Lipschitz and growth
assumptions that ensure existence and strong uniqueness of the solutions
(see for example Theorem 5.2.9 of Karatzas and Shreve [5])
le-logo
Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

6 / 32

We have M coupled SDEs and M + 1 noise sources.

The dynamics of a given stock depends on all the other stocks
composing the index through the volatility term (t, ItM ).
The cross-correlations between stocks are not constant but stochastic :
ij =

i j 2 (t, ItM )
i2 2 (t, ItM ) + i2 (t, Sti,M )

j2 2 (t, ItM ) + j2 (t, Stj,M )

Note that they depend not only on the stocks but also on the index.

le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

7 / 32

The index I M satisfies the following SDE

dItM = rItM dt
+

j,M
M
j=1 j wj St

dt +

j,M
M
j=1 j wj St

(t, ItM )dBt

j,M
j,M
j
M
j=1 wj St j (t, St )dWt

(2)

Our model is inline with Cizeau et al. [2] :

The beta coefficients are narrowly distributed around 1
j,M
M

ItM .
j=1 j wj St
For large M, we will show that the term
neglected.

j
j
j
M
j=1 wj St j (t, St )dWt

can be

= rj = j rI M + j W j + drift
where rj (resp. rI M ) is the log-return of the stock j (resp. the index).
The return of a stock is decomposed into a systemic part driven by the index,
le-logo
which represents the market, and a residual part.
Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

8 / 32

A simplified Model
We look at the asymptotics for a large number of underlying stocks.
Consider the limit candidate (It )t[0,T] solution of
I0 = I0M

dIt /It = (r )dt + (t, It )dBt ;

(3)

Theorem 1
Let p . If
(H1) Kb s.t. (t, s), |(t, s)| + |j (t, s)| Kb
K s.t. (t, s1 , s2 ), |s1 (t, s1 ) s2 (t, s2 )| K |s1 s2 |.
then, CT a constant independent of M such that

sup |ItM It |2p

0tT

CT

M
2
j=1 wj

+
Sbai (UPE-CERMICS)

Coupling Index and Stocks

M
j=1 wj |j

M
j=1 wj |j

2p

2p
le-logo

CNF 2009

9 / 32

We control the error when we replace I M by I in the dynamics of the stocks :

Theorem 2
M

j
Denote by I t = M
j=1 wj St . Under the assumptions of Theorem 1 and if
(H2) K s.t. (t, s1 , s2 ), |s1 (t, s1 ) s2 (t, s2 )| K |s1 s2 |
KLip s.t. (t, s1 , s2 ), |(t, s1 ) (t, s2 )| KLip |s1 s2 |
then, j {1, . . . , M}, CTj s.t.

sup |Stj,M Stj |2p

0tT

CTj

M
2
j=1 wj

sup |ItM I t |2p

0tT

Sbai (UPE-CERMICS)

M
j=1 wj |j

max1jM CTj
+

M
j=1 wj |j

Coupling Index and Stocks

2p

2p

M
j=1 wj

M
j=1 wj |j

2p

2p

M
2
j=1 wj
M
j=1 wj |j

CNF 2009

2p
le-logo

10 / 32

The limit M +
Corollary
Under the additional assumptions
(H3) A s.t. maxj1 (S0j,M )2 + (jM )2 + (jM )2 A,
M
M 2
j=1 (wj ) M 0,
M
M M
j=1 wj |j | M 0,

(H4) PM
w =
(H5) PM
=
M

M
wM
j |j | 0,

(H6) PM
=
j=1

one has sup0tT |ItM It |2 M 0 and

j {1, . . . , M},
M

sup0tT |Stj,M Stj |2 M 0. If, in addition,

2
M
wM
M 0.
j < then sup0tT |It I t |
M

sup
M j=1

le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

11 / 32

Assumption (H4) prevents concentration in the weights (for example

M
1
1
uniform weights lead to PM
w =
j=1 M 2 = M M 0).
In order to remain coherent with the definition of the beta coefficients,
we have to take = 1.

PM
w
0.026

opt
0.975

2
(PM
opt )
0.0173

2
(PM
=1 )
0.0174

M 2
TABLE: Computation of PM
w , opt and (Popt ) for the Eurostoxx index at December
21, 2007. The beta coefficients are estimated on a two year history.

le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

12 / 32

To sum up, under mild assumptions, when the number of underlying stocks is
large, our original model may be approximated by

j {1, . . . , M},

dStj

= (r j )dt + j (t, It )dBt + j (t, Stj )dWtj

Stj
dIt
= (r I )dt + (t, It )dBt .
It

(4)

We end up with
A local volatility model for the index
A stochastic volatility model for each stock, decomposed into a systemic
part driven by the index level and an intrinsic part.
Careful ! Our simplified model is not valid for options written on the index
together with all its composing stocks since the index is no longer an exact,
but an approximate, weighted sum of the stocks. Instead, one should consider
M
j
le-logo
the reconstructed index I t = M
j=1 wj St or use the original model.
Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

13 / 32

Outline

Introduction

Model Specification

Calibration

Numerical experiments

Conclusion

le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

14 / 32

Calibration of the simplified model

dSt
= (r )dt + (t, It )dBt + (t, St )dWt
St

(5)

dIt
= (r I )dt + (t, It )dBt .
It

Fitting the index smile boils down to the calibration of a local volatility
model.
Fitting an individual stock smile is more tedious.
Our model gives an advantage to the fit of index option prices (index
options are usually more liquid than individual stock options).

le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

15 / 32

Non-parametric estimation of
We have a relation between the local volatility and the stochastic volatility
(see Gyongi [4] or Dupire [3]) :
vloc (t, K) = 2 (t, K) + 2 2 (t, It ) | St = K
So,
(t, K) =

vloc (t, K) 2 ( 2 (t, It ) | St = K).

(6)

vloc can be calibrated with the best-fit of a parametric form to the stock
market smile.
Estimating the conditional expectation is more challenging (it depends
implicitly on as it is the case for (St , It )).
We investigate a simulation based approach yielding a non-parametric
estimation of .
le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

16 / 32

If we plug the formula (6) in the dynamics of the stock we obtain

dSt
= (r )dt + (t, It )dBt +
St

vloc (t, St ) 2 ( 2 (t, It ) | St )dWt

dIt
= (r I )dt + (t, It )dBt
It
This SDE is non-linear in the sense of McKean.
Kernel estimators of the Nadaraya-Watson type :
N

2 (t, Iti )K

2 (t, It ) | St = s

i=1
N

K
i=1

s Sti
hN

s Sti
hN

where K is a non-negative kernel s.t. K(x)dx = 1 and lim hN = 0.le-logo

N
Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

17 / 32

A system of interacting particles

Replacing the conditional expectation by its non-parametric estimator yield
the following system : 1 i N,
dSti,N
Sti,N
dIti
Iti

= (r )dt+ (t, Iti )dBit +

vloc (t, Sti,N )

N
j=1

2 (t,Itj )K
N
j=1

j,N
i,N
St St
hN

j,N
i,N
St St
hN

dWti

= (r I )dt + (t, Iti )dBit

(Bi , W i )i1 is a sequence of independent two-dimensional Brownian motions.

This 2N-dimensional linear SDE may be discretized using a simple Euler
scheme !
le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

18 / 32

Calibration of the original model

j {1, . . . , M},

dStj,M
Stj,M

= (r j )dt + j (t, ItM )dBt + j (t, Stj,M )dWtj

M

wi Sti,M

ItM =
i=1

A perfect calibration of both the index and the individual stocks is

complicated... but we can
take for the calibrated local vol of the index and then calibrate the j
coefficients in order to fit all the individual stock smiles the index is not
perfectly calibrated but the error should be small (Theorem 1).
take for and j the calibrated coefficients in the simplified model the
index and the stocks are not perfectly calibrated but the error should be
small (Theorems 1 and 2).
le-logo
We allow for a slight error in the calibration but the additivity constraint
is observed.
Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

19 / 32

Outline

Introduction

Model Specification

Calibration

Numerical experiments

Conclusion

le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

20 / 32

An acceleration technique
The simulation of the particle system is time consuming : a global
complexity of order O(nN 2 ) where n is the number of time steps in the
Euler scheme.
A possible acceleration technique : neglect particles which are far away
from each other.
How ? Sort the particles and stop the estimation of the conditional
expectation whenever the contribution of a particle is lower than some
fixed threshold.
We lose in precision but we gain much more in computation time.
le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

21 / 32

Data :
Local volatilities of the Eurostoxx index and of Carrefour at December
21, 2007.
Beta coefficient estimated on a two years history ( = 0.7).
Short interest rate and dividend yields as of December 21, 2007.
Maturity T = 1.
Threshold for the accelerated technique :

1
N.

Smoothing parameter : hN = N 10 .
Number of time steps for the Euler scheme : n = 20.
Moneyness ( SK0 )
Error : |simul exact |

0.5
36

0.7
8

0.9
2

1
1

1.1
2

1.2
9

1.5
32

2
56

TABLE: Error (in bp) on the implied volatility with N = 200000 particles.
le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

22 / 32

0.40
Exact Implied Vol.
N=10000

0.38

N=200000
0.36

0.34

0.32

0.30

0.28

0.26

0.24
0.5

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

1.4

1.5

Moneyness

F IGURE: Convergence of the implied volatility obtained with non-parametric

estimation.

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

le-logo

23 / 32

Illustration of theorems 1 and 2

1

The original model

j {1, . . . , M},

dStj,M
= rdt + (t, ItM )dBt
Stj,M
i,M
ItM = M
i=1 wi St .

+ (t, Stj,M )dWtj

The simplified model

j {1, . . . , M},

dStj
Stj
dIt
It

= rdt + (t, It )dBt + (t, Stj )dWtj

= rdt + (t, It )dBt .

i
Reconstructed index I t = M
i=1 wi St .
The constant-correlation model

j {1, . . . , M},

dStj
Stj

= rdt +

vloc (t, Stj )dWtj

i = j, d < W i , W j >t = dt.

Sbai (UPE-CERMICS)

Coupling Index and Stocks

le-logo

CNF 2009

24 / 32

We take for the calibrated local vol of the Eurostoxx.

We choose an arbitrary parametric form for the vol coefficient .
We evaluate vloc s.t. the constant-correl model and the simplified one
yield the same implied vol for individual stocks
(vloc (t, S) = 2 (t, S) + ( 2 (t, It )|St = S)).
We fix the correlation coefficient s.t. the constant-correl model and the
simplified one yield the same ATM implied vol for the index.
We take the same market data as before.

le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

25 / 32

0.42
Simplified
0.40

Market
Original

0.38
0.36
0.34
0.32
0.30
0.28
0.26
0.24
0.22
0.20
0.5

1.0

1.5

2.0

Moneyness

F IGURE: Implied volatility of an individual stock.

le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

26 / 32

0.40
Simplified
Market
Original
0.35

Simplified Reconstructed

0.30

0.25

0.20

0.15
0.5

1.0

1.5

2.0

Moneyness

F IGURE: Implied volatility of the index.

le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

27 / 32

Outline

Introduction

Model Specification

Calibration

Numerical experiments

Conclusion

le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

28 / 32

We have introduced a new model for describing the joint evolution of an

index and its composing stocks.
The index induces some feedback on the dynamics of its stocks.
For large number of underlying stocks, the model reduces to a local vol
model for the index and to a stochastic vol for each individual stock with
volatility driven by the index.
We favor the fit of the index smile.
We have proposed a simulation based approach allowing to fit both the
index and the stocks smiles.

le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

29 / 32

Thank you !

le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

30 / 32

References I
M. Avellaneda, D. Boyer-Olson, J. Busca, and P. Friz.
Reconstructing volatility.
Risk, pages 8791, October 2002.
P. Cizeau, M. Potters, and J-P. Bouchaud.
Correlation structure of extreme stock returns.
Quantitative Finance, 1(2) :217222, February 2001.
B. Dupire.
Pricing with a smile.
Risk, pages 1820, January 1994.
I. Gyongy.
Mimicking the one-dimensional marginal distributions of processes
having an Ito differential.
Probab. Theory Relat. Fields, 71(4) :501516, 1986.
le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

31 / 32

References II

I. Karatzas and S. E. Shreve.

Brownian motion and stochastic calculus.
Springer-Verlag New-York, second edition, 1991.
P. Lee, L. Wang, and A. Kerim.
Index volatility surface via moment-matching techniques.
Risk, pages 8589, December 2003.

le-logo

Sbai (UPE-CERMICS)

Coupling Index and Stocks

CNF 2009

32 / 32