Risk Neutralmodeltohedgingandpricinpowerderivatives Seminar

Position of the problem
Spot model
Pricing & hedging
Conclusion
A structural risk-neutral model
for pricing and hedging power derivatives
Rene Ad, Luciano Campi, Nicolas Langrene
Universities Paris 13 & Paris 7
EDF R&D - FiME Research Centre
Rene Ad, Luciano Campi, Nicolas Langrene Universities Paris 13 & Paris 7 EDF R&D - FiME Research Centre A structural risk-neutral model for pricing and hedging power derivatives 1 / 45
Spot model
Pricing & hedging
Conclusion
Contents
1
Electricity prices modeling
Related works
2
Spot model
Design
Estimation
3
Pricing & hedging
Dynamics of fuels etc
Local risk minimization
Futures
Options
Hedging with futures on electricity
4
Conclusion
Spot model
Pricing & hedging
Conclusion
Related works
Looking for a power spot price model
Applications
pricing of derivatives on the spot
asset valuation (strip of hourly fuel spread options)
hedging
energy market risk management
Model requirements
realistic
robust
tractable
consistent
Spot model
Pricing & hedging
Conclusion
Related works
Two types of modeling
Modeling futures prices
pros modeling the real available instruments
cons introduction of many parameters to reconstruct
hourly futures prices
Modeling spot prices
1
Exogeneous
pros tractability
cons correlation
2
Equilibrium
pros correlation
cons complexity
Spot model
Pricing & hedging
Conclusion
Related works
Related works
Electricity prices exogeneous dynamics
Deng (00), Benth et al. (03, 07, 09), Burger et al. (04), Kolodnyi
(04), Cartea & Figueroa (05), Geman & Roncoroni (06)
Equilibrium model
Spot Futures Options
Pirrong & Jermakyan (00)
Barlow (02)
Kanamura & Ohashi (07)
Cartea & Villaplana (08)
Coulon & Howison (09)
Lyle & Elliot (09)
Spot model
Pricing & hedging
Conclusion
Related works
This talk
Objectives
pricing and hedging power derivatives...
... using an improved version of Ad, C., Nguyen & Touzi
(09) Structural Risk-Neutral model
Spot Futures Options
Ad, C., Nguyen & Touzi (09)
improved SRN model
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Initial SRN Model
Variables
n fuels, 1 i n
D
t
demand (MW)
C
i
t
capacities (en MW)
S
i
t
fuel prices
h
i
heat rates (h
i
S
i
t
en e/MWh, en i )
Electricity price (e/MWh)
P
t
=
n
i =1
h
i
S
i
t
1
{
i 1
k=1
C
k
t
D
t
i
k=1
C
k
t
}
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Initial SRN Model
Variables
n fuels, 1 i n
D
t
demand (MW)
C
i
t
capacities (en MW)
S
i
t
fuel prices
h
i
heat rates (h
i
S
i
t
en e/MWh, en i )
P
t
=
n
i =1
h
i
S
i
t
1
{
i 1
k=1
C
k
t
D
t
i
k=1
C
k
t
}
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Initial SRN Model
Variables
n fuels, 1 i n
D
t
demand (MW)
C
i
t
capacities (en MW)
S
i
t
fuel prices
h
i
heat rates (h
i
S
i
t
en e/MWh, en i )
P
t
=
n
i =1
h
i
S
i
t
1
{
i 1
k=1
C
k
t
D
t
i
k=1
C
k
t
}
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Initial SRN Model
Variables
n fuels, 1 i n
D
t
demand (MW)
C
i
t
capacities (en MW)
S
i
t
fuel prices
h
i
heat rates (h
i
S
i
t
en e/MWh, en i )
P
t
=
n
i =1
h
i
S
i
t
1
{
i 1
k=1
C
k
t
D
t
i
k=1
C
k
t
}
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Initial SRN Model
Variables
n fuels, 1 i n
D
t
demand (MW)
C
i
t
capacities (en MW)
S
i
t
fuel prices
h
i
heat rates (h
i
S
i
t
en e/MWh, en i )
P
t
=
n
i =1
h
i
S
i
t
1
{
i 1
k=1
C
k
t
D
t
i
k=1
C
k
t
}
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Initial SRN Model
Variables
n fuels, 1 i n
D
t
demand (MW)
C
i
t
capacities (en MW)
S
i
t
fuel prices
h
i
heat rates (h
i
S
i
t
en e/MWh, en i )
P
t
=
n
i =1
h
i
S
i
t
1
{
i 1
k=1
C
k
t
D
t
i
k=1
C
k
t
}
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Initial SRN Model
Variables
n fuels, 1 i n
D
t
demand (MW)
C
i
t
capacities (en MW)
S
i
t
fuel prices
h
i
heat rates (h
i
S
i
t
en e/MWh, en i )
P
t
=
n
i =1
h
i
S
i
t
1
{
i 1
k=1
C
k
t
D
t
i
k=1
C
k
t
}
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Initial SRN model
Pros
Consistency between electricity prices and fuel prices
Consistency between electricity prices and demand
Cons
Marginal fuel cost is not the spot price
1
Non-convex technical constraints
2
Strategic behaviour (Horta csu & Puller, RAND J. of
Economics 2008)
3
Fixed cost recovery problem for peak-load generation plants
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Initial SRN model
Pros
Consistency between electricity prices and fuel prices
Consistency between electricity prices and demand
Cons
Marginal fuel cost is not the spot price
1
Non-convex technical constraints
2
Strategic behaviour (Horta csu & Puller, RAND J. of
Economics 2008)
3
Fixed cost recovery problem for peak-load generation plants
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Initial SRN Model - illustration
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Improved SRN model
Marginal fuel cost

P
t
:=
n
i =1
h
i
S
i
t
1
{
i 1
k=1
C
k
t
D
t
i
k=1
C
k
t
}
Available capacity C
t
:=
n
k=1
C
k
t
Price spikes occur when the electric system is under stress, i.e.
C
t
D
t
is small
y
t
:=
P
t
P
t
as a (nonlinear) function of x
t
:= C
t
D
t
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Improved SRN model
Marginal fuel cost

P
t
:=
n
i =1
h
i
S
i
t
1
{
i 1
k=1
C
k
t
D
t
i
k=1
C
k
t
}
t
:=
n
k=1
C
k
t
C
t
D
t
is small
y
t
:=
P
t
P
t
t
:= C
t
D
t
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Improved SRN model
Marginal fuel cost

P
t
:=
n
i =1
h
i
S
i
t
1
{
i 1
k=1
C
k
t
D
t
i
k=1
C
k
t
}
t
:=
n
k=1
C
k
t
C
t
D
t
is small
y
t
:=
P
t
P
t
t
:= C
t
D
t
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Improved SRN model
Marginal fuel cost

P
t
:=
n
i =1
h
i
S
i
t
1
{
i 1
k=1
C
k
t
D
t
i
k=1
C
k
t
}
t
:=
n
k=1
C
k
t
C
t
D
t
is small
y
t
:=
P
t
P
t
t
:= C
t
D
t
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Improved SRN model - Estimation
Figure: PowerNext - 19th hours
Nov, 13th 06 to April 30th 10
Observation
Decreasing relation
Dicult estimation
Idea
Quantiles
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Observation
Decreasing relation
Dicult estimation
Idea
Quantiles
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Observation
Decreasing relation
Dicult estimation
Idea
Quantiles
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Estimated relation : y
t
=

x
t
Improved SRN model
P
t
= g
_
n
k=1
C
k
t
D
t
_
_
n
i =1
h
i
S
i
t
1
{
i 1
k=1
C
k
t
D
t
i
k=1
C
k
t
}
_
with scarcity function
g (x) := min
_

x
, M
_
1
{x>0}
+ M1
{x0}
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Improved SRN model - Back-testing
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Improved SRN model - Back-testing
Spot model
Pricing & hedging
Conclusion
Design
Estimation
Improved SRN model - Backtesting
Spot model
Pricing & hedging
Conclusion
Futures
Options
Pricing & hedging
Pricing
incomplete market
need for a hedging criterion
super-replication, utility indierence or mean-variance
our choice : Local Risk Minimization
Local Risk Minimization (Pham (00), Schweizer (01))
valuation : expected discounted payo under

Q
allows to decompose contingent claim into hedgeable part
(fuels) and non-hedgeable part (demand, capacities)
allows explicit formulas
Spot model
Pricing & hedging
Conclusion
Futures
Options
Pricing & hedging
Pricing
incomplete market
need for a hedging criterion
super-replication, utility indierence or mean-variance
our choice : Local Risk Minimization
Local Risk Minimization (Pham (00), Schweizer (01))
valuation : expected discounted payo under

Q
allows to decompose contingent claim into hedgeable part
(fuels) and non-hedgeable part (demand, capacities)
allows explicit formulas
Spot model
Pricing & hedging
Conclusion
Futures
Options
Dynamics of fuels
Assume r constant, convenience yields and storage costs zero for
simplicity.
Fuels
n 1 fuels (as coal, gas, oil ...) whose cost h
i
S
i
to produce 1
MWh of electricity follows
dS
i
t
= S
i
t
(
i
t
dt +
i
t
dW
S,i
t
)
where W
S,i
are correlated BMs and coes are chosen so that
h
1
S
1
< . . . < h
n
S
n
(model spreads Y
i
= h
i +1
S
i +1
h
i
S
i
as
independent geometric BMs).
NA and completeness assumption
There exists a unique risk-neutral probability Q P for S.
Spot model
Pricing & hedging
Conclusion
Futures
Options
Local risk minimization I
Roughly speaking, let X be a multidimensional (discounted) price
process
Introduced by Follmer-Schweizer (1991)
Under regularity condition, any payo H with maturity T
H = H
0
+
_
T
0
H
t
dX
t
+ L
H
T
where H
0
R and L
H
martingale orthogonal to X.
_
T
0
dX is the hedgeable part, L
H
T
the residual risk, H
0
+ L
H
T
the cost of the strategy
How to compute H
0
,
H
, L
H
? Easy when X is a martingale :
use Kunita-Watanabe decomposition !
Spot model
Pricing & hedging
Conclusion
Futures
Options
process
H = H
0
+
_
T
0
H
t
dX
t
+ L
H
T
where H
0
R and L
H
_
T
0
H
T
0
+ L
H
T
How to compute H
0
,
H
, L
H
Spot model
Pricing & hedging
Conclusion
Futures
Options
process
H = H
0
+
_
T
0
H
t
dX
t
+ L
H
T
where H
0
R and L
H
_
T
0
H
T
0
+ L
H
T
How to compute H
0
,
H
, L
H
Spot model
Pricing & hedging
Conclusion
Futures
Options
process
H = H
0
+
_
T
0
H
t
dX
t
+ L
H
T
where H
0
R and L
H
_
T
0
H
T
0
+ L
H
T
How to compute H
0
,
H
, L
H
Spot model
Pricing & hedging
Conclusion
Futures
Options
Local risk minimization II
When X is not a martingale but not far from being so ...
Follmer-Schweizer (1991) : there exists a risk-neutral

Q for X
s.t.
H =

E[H] +
_
T
0
H
t
dX
t
+
L
H
T
H
0
=

E[H],
H
=

H
, L
H
=

L
H
Q is called minimal equivalent martingale measure

Hobson (2005) : in diusion models with non-tradable assets,
E[H] is an upper bound for U-indierence bid price.

Spot model
Pricing & hedging
Conclusion
Futures
Options

Q for X
s.t.
H =

E[H] +
_
T
0
H
t
dX
t
+
L
H
T
H
0
=

E[H],
H
=

H
, L
H
=

L
H


Spot model
Pricing & hedging
Conclusion
Futures
Options

Q for X
s.t.
H =

E[H] +
_
T
0
H
t
dX
t
+
L
H
T
H
0
=

E[H],
H
=

H
, L
H
=

L
H


Spot model
Pricing & hedging
Conclusion
Futures
Options

Q for X
s.t.
H =

E[H] +
_
T
0
H
t
dX
t
+
L
H
T
H
0
=

E[H],
H
=

H
, L
H
=

L
H


Spot model
Pricing & hedging
Conclusion
Futures
Options

Q for X
s.t.
H =

E[H] +
_
T
0
H
t
dX
t
+
L
H
T
H
0
=

E[H],
H
=

H
, L
H
=

L
H


Spot model
Pricing & hedging
Conclusion
Futures
Options
Dynamics of demand and capacities
Demand & capacities
dD
t
= a (t, D
t
) dt + b (t, D
t
) dW
D
t
dC
i
t
=
i
_
t, C
i
t
_
dt +
i
_
t, C
i
t
_
dW
C,i
t
where W
D
, W
C
and W
S
are independent BMs.
Minimal martingale measure
The minimal martingale measure

Q is given by
Q|
F
t
= Q|
F
S
t
P|
F
C,D
t
where F
t
is the ltration generated by W
S
, W
D
, W
C
.
Spot model
Pricing & hedging
Conclusion
Futures
Options
Futures
Under our assumptions, we can prove the following
Futures prices F
e
t
(T) =

E
t
[P
T
]
F
e
t
(T) =
n
i =1
h
i
G
T
i
(t, C
t
, D
t
) F
i
t
(T)
with :
G
T
i
(t,C
t
,D
t
) = E
t
_
g
_
n
k=1
C
k
T
D
T
_
1
{
i 1
k=1
C
k
T
D
T

i
k=1
C
k
T
}
_
Spot model
Pricing & hedging
Conclusion
Futures
Options
Futures prices - hedging
Futures price dynamics
dF
e
t
(T) =
n
i =1
h
i
_
G
T
i
(t, C
t
, D
t
)dF
i
t
(T) + F
i
t
(T)dG
T
i
(t, C
t
, D
t
)
dG
T
i
(t, C
t
, D
t
) =
n
k=1
G
T
i
c
k
(t, C
t
, D
t
)
k
(t, C
k
t
)dW
C,k
t
+
G
T
i
z
(t, C
t
, D
t
)b(t, D
t
)dW
D
t
so that
dF
e
t
(T) =
S
t
dW
t
+
C
t
dW
C
t
+
D
t
dW
D
t
for adapted suitable processes
S
,
C
,
D
, which are explicitly
computable.
Spot model
Pricing & hedging
Conclusion
Futures
Options
To go further, need to choose more specic dynamics for
demand and capacities
deterministic part for seasonality + Ornstein-Uhlenbeck
G
T
i
explicite as function of extended incomplete
Goodwin-Staton integral :
G (x, y; ) =
_

x
1
(y + z)
e
z
2
dz
... for which ecient numerical algorithms are provided in Ad,
C. & Langrene (10).
Spot model
Pricing & hedging
Conclusion
Futures
Options
Futures prices - hedging : spot simulations
Spot model
Pricing & hedging
Conclusion
Futures
Options
Spot model
Pricing & hedging
Conclusion
Futures
Options
Spot model
Pricing & hedging
Conclusion
Futures
Options
Spot model
Pricing & hedging
Conclusion
Futures
Options
Spot model
Pricing & hedging
Conclusion
Futures
Options
Numerical test
Hedging a 3-months
electricity futures with a
delivery period of 1 hour
with a daily rebalanced
basket of futures
contracts on fuels
Spot model
Pricing & hedging
Conclusion
Futures
Options
Numerical test
Hedging a 3-months
electricity futures with a
delivery period of 1 hour
with a daily rebalanced
basket of futures
contracts on fuels
Spot model
Pricing & hedging
Conclusion
Futures
Options
Remarks
Positive values are
losses
Far from maturity :
perfect hedge ;
electricity futures is
equivalent to a
basket of fuels
Close to maturity :
inecient hedge
Spot model
Pricing & hedging
Conclusion
Futures
Options
Spread options
Spread option with a 2 fuel model
The price
0
at time t = 0 of a call spread option with pay-o
H = (P
T
h
1
S
1
T
K)
+
is given by :
0
=
R
2
f
C
1
T
D
T
(z)f
C
2
T
(c)
1
(c, z)1
{z>0}
+
2
(c, z)1
{z0}
dcdz,
1
= (g 1)BS
0
(
1
, K)1
{g>1}
2
= g
f
Y
1
T
(y)BS
0
2
,
K + (1 g)y
g
1
{g1}
+ 1
{g>1}
1
{y<
K
g1
}
dy
+
gY
2
0
N
r

2
1
2
T ln
K
(g1)Y
1
0
+ (g 1) BS
0
1
,
K
g 1
1
{g>1}
with g := g(c + z).
Spot model
Pricing & hedging
Conclusion
Futures
Options
Spread options
semi-explicit formula : numerical integration
partial hedging with futures on fuels and electricity,
semi-explicit formulae for partial hedging strategy (not only
for spread options)
applied on European dark spread (i.e. energy - gas) call option
with a period of delivery of 1 hour
Spot model
Pricing & hedging
Conclusion
Futures
Options
Consider H = (F
e
T
(T
), F
T
(T
), C
T
, D
T
) with T
> T.
By Markov, its

Q-price in t is (t, F
t
, C
t
, D
t
) with (t, x, c, z)
regular
Hs decomposition into hedgeable part/residual risk is
H =

E[H] +
_
T
0
t
dF
t
+
_
T
0
e
t
dF
e
t
+
L
T
where
e
t
=
1
||(
C
t
,
D
t
)||
2
_
C,i
t

i
c
i
+
D
t
b
z
i
t
=
y
i
+
h
i
G
T
i
||(
C
t
,
D
t
)||
2
_
C,i
t

i
c
i
+
D
t
b
z
L
T
can be computed explicitly as well
Spot model
Pricing & hedging
Conclusion
Futures
Options
Consider H = (F
e
T
(T
), F
T
(T
), C
T
, D
T
) with T
> T.
By Markov, its

Q-price in t is (t, F
t
, C
t
, D
t
) with (t, x, c, z)
regular
Hs decomposition into hedgeable part/residual risk is
H =

E[H] +
_
T
0
t
dF
t
+
_
T
0
e
t
dF
e
t
+
L
T
where
e
t
=
1
||(
C
t
,
D
t
)||
2
_
C,i
t

i
c
i
+
D
t
b
z
i
t
=
y
i
+
h
i
G
T
i
||(
C
t
,
D
t
)||
2
_
C,i
t

i
c
i
+
D
t
b
z
L
T
can be computed explicitly as well
Spot model
Pricing & hedging
Conclusion
Conclusion
Conclusions
SRN electricity spot price model with a scarcity function
allows futures and derivatives pricing and hedging
nevertheless, only fuels dependent part can be hedged ...
... unless we use energy future for partially hedging demand
and capacities risk (see the paper Ad-C.-Langrene)
Perspectives
comparison with realquoted futures, calibration dynamics
utility-based pricing of futures, options ...
optimal investment/production problem, optimal switching
Spot model
Pricing & hedging
Conclusion
Conclusion
Conclusions
SRN electricity spot price model with a scarcity function
allows futures and derivatives pricing and hedging
nevertheless, only fuels dependent part can be hedged ...
... unless we use energy future for partially hedging demand
and capacities risk (see the paper Ad-C.-Langrene)
Perspectives
comparison with realquoted futures, calibration dynamics
utility-based pricing of futures, options ...
optimal investment/production problem, optimal switching
Spot model
Pricing & hedging
Conclusion
References
Ad, C., Nguyen Huu & Touzi, Int. J. Theoretical & Applied
Finance, 2010
Ad, C., Langrene, 2010, available on HAL
Barlow, Math. Finance, 2002
Benth & Koekebakker, J. of Energy Economics, 2007
Benth & Vos, Tech. Rep., Math. Dept. Oslo, 2009
Benth, Ekeland, Hauge & Nielsen, Applied Math. Finance,
2003
Burger, Klar, M uller & Schlindlmayr, Quantitative Finance,
2004
Spot model
Pricing & hedging
Conclusion
References
Cartea & Figueroa, Applied Math. Finance, 2005
Cartea & Villaplana, J. of Banking & Finance, 2008
Coulon & Howison, J. of Energy Markets, 2009
Deng, Tech. Rept.,California Energy Institute, 2000
Geman & Roncoroni, J. of Business, 2006
Kanamura & Ohashi, Energy Economics, 2007
Kolodnyi, J. of Engineering Mathematics, 2004
Spot model
Pricing & hedging
Conclusion
References
Lyle & Elliott, Energy Economics, 2009
Pham, Math. Meth. of Operations Research, 2000
Pirrong & Jermakyan, J. of Banking & Finance, 2008
1
Schweizer, Handbook Math. Finance, Cambridge Univ. Press,
2001
1. Olin Business School Tech. Rep. 2000

Risk Neutralmodeltohedgingandpricinpowerderivatives Seminar

Transféré par

Informations du document

Description originale:

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Risk Neutralmodeltohedgingandpricinpowerderivatives Seminar

Transféré par

Droits d'auteur :

Formats disponibles

Position of the problem

Q is called minimal equivalent martingale measure

E[H] is an upper bound for U-indierence bid price.

Q is called minimal equivalent martingale measure

E[H] is an upper bound for U-indierence bid price.

Q is called minimal equivalent martingale measure

E[H] is an upper bound for U-indierence bid price.

Q is called minimal equivalent martingale measure

E[H] is an upper bound for U-indierence bid price.

Q is called minimal equivalent martingale measure

E[H] is an upper bound for U-indierence bid price.

Vous aimerez peut-être aussi