Strap?

l Cutting edge

Corridor variance swaps

Peter Carr and Keith Lewis

I
t is widely recognised that delta-hedged positions in options can be used a Monte Carlo simulation of our payout definition and hedging strategy.
to trade volatility. To facilitate volatility trading for their clients, several The penultimate section discusses corridor variance swaps when the cor-
institutions routinely offer variance swaps. A variance swap is a finan- ridor is defined by two positive finite constants. A final section summaris-
cial contract that upon expiry pays the difference between a standard his- es and suggests extensions.
torical estimate of daily return variance and a fixed rate determined at
inception. As in any swap, the fixed rate is initially chosen so that a vari- Structuring upside and downside variance swaps
ance swap has zero cost to enter. See Demeterfi et al (1999) or Carr & Here, we define the payouts to upside and downside variance swaps. For
Madan (1998) for pricing and see Chriss & Morokoff (1999) for risk man- an upside variance swap, the corridor needed to define the payout is the
agement issues. semi-infinite interval (L, ∞), where the fixed constant L ≥ 0 denotes the
Over the past few years, several institutions have also begun offering lower bound. Upside variance swaps can be used to create other corridor
corridor variance swaps. These differ from standard variance swaps only variance swaps. For example, when L = 0, the payout to an upside vari-
in that the underlying’s price must be inside a specified corridor in order ance swap will degenerate into the payout from a standard variance swap.
for its squared return to be included in the floating part of the variance For a downside variance swap, the payout will be given by the difference
swap payout. As in a standard variance swap, the fixed payment is made between the payout to a standard variance swap and an upside variance
at maturity and is initially chosen so that the corridor variance swap has swap. To create a corridor variance swap whose supporting corridor has
zero cost to enter. In the corridor variance swap considered in this article, a positive lower bound and a finite upper bound, we can take the differ-
the fixed payment is independent of the occupation time of the corridor. ence of two upside variance swaps with different lower bounds, as will be
However, variations exist in which the fixed payment accrues over time at discussed in the penultimate section.
a constant rate only while the underlying is in the corridor. Consider a finite set of discrete times {t0, t1, ... , tn} at which the path
A corridor variance swap is a generalisation of a standard variance swap of some underlying is monitored. In what follows, we will use a futures
in that the latter results from the former when the corridor is extended to price as the underlying and we will take the monitoring times to be daily
all possible price levels. An upside variance swap uses a corridor extend- closings. Let F0 denote the known initial futures price and let Fi ≥ 0 de-
ing from a fixed barrier up to infinity, while a downside variance swap note the random futures price at the close of day i, for i = 1, 2, ... , n. For
uses a corridor extending from a fixed barrier down to zero. From the spec- an upside variance swap, the futures price is said to start in the corridor
ulator’s perspective, the advantage of a corridor variance swap over a vari- on day i if Fi – 1 > L and it is said to stop in the corridor on day i if Fi > L.
ance swap is that it allows the expression of a view on volatility that is The opposite inequalities hold for downside variance swaps. For an up-
contingent upon the price level. For example, an investor who thinks that side variance swap, the futures price is said to enter the corridor on day i
returns are likely to be more negatively skewed than predicted by the mar- if Fi – 1 ≤ L and Fi > L. In contrast, it is said to exit the corridor on day i if
ket might buy a downside variance swap and sell an upside variance swap. Fi – 1 > L and Fi ≤ L. For a downside variance swap, entry occurs on days
From the hedger’s perspective, the advantage of a corridor variance swap when the futures price exits the upside corridor. Likewise, exit occurs on
over a standard variance swap is that the hedge involves fewer initial po- days when the futures price enters the upside corridor.
sitions and less frequent revision over time. The exact specification of the payout to a corridor variance swap dif-
Carr & Madan (1998) show how to synthesise continuously monitored fers from firm to firm. Our specification of the payout to a corridor vari-
variance and corridor variance swaps when the underlying price process ance swap is chosen so that the hedging error can be made third order
is assumed to be continuous. The purpose of this article is to show how without imposing a model for price dynamics. We also insist that the
to accurately approximate the payout to discretely monitored variance and payouts to upside and downside variance swaps be defined so that they
corridor variance swaps under no assumptions about the underlying price sum to the payout of a standard variance swap. To begin specifying the
process. Given the increasing recognition of the importance of jumps and payout of a corridor variance swap, let 1Fi – 1∈Ri – 1,Fi∈Ri denote the indi-
that all swaps are monitored daily in practice, these extensions are long cator function, which is one when Fi – 1 is in region Ri – 1 and Fi is in re-
overdue. We show that with frictionless markets and deterministic interest gion Ri, but is zero otherwise. An upside variance contract is defined to
rates, the payout to a corridor variance swap can be accurately approxi- be a financial security that has the following non-negative payout at the
mated by combining at most daily trading in the underlying with static po- fixed time tn:
sitions in standard European-style options maturing with the swap and
struck inside the corridor. In particular, the approximation error is third n 
2 2
  F   F
order and hence the strategy replicates well if third and higher powers of Qnu ( L) ≡ ∑ 1Fi−1 > L, Fi > L  ln i  + 1Fi−1 ≤ L, Fi > L  ln i 
i =1   Fi −1   L
daily returns sum to a negligible amount. 
The structure of this article is as follows. The next section defines the
 F  2  F  2   (1)
payouts to upside and downside variance swaps. The following section + 1Fi−1 > L, Fi ≤ L  ln i  −  ln i   
shows how to approximate the payouts to these swaps by combining sta-  Fi −1   L  
 
tic positions in options struck inside the corridor with at most daily trad-
ing in the underlying futures. The subsequent section shows the results of The first term in the summand is due to days in which the futures price

WWW.RISK.NET ● FEBRUARY 2004 RISK XX

Cutting edge l Strap?

starts and stops inside the upper corridor, while the last two terms are By definition, a corridor variance swap is a swap with a single payment
due to entry and exit of the corridor respectively. If the futures price starts at maturity given by the difference between a floating part and a fixed part:
and stops below the upper corridor on day i, then that day’s move is ig-
nored. If the futures price starts and stops in the corridor on day i, then VSnc ( L) = Qnc ( L) − F0c ( L) (6)
that day’s percentage change is squared. If the futures price enters the
corridor on day i, then only the percentage change from L is squared. If The floating part is the payout Qnc(L) to a corridor variance contract, where
the futures price exits the corridor on day i, then the square of the per- the superscript c takes on the value u for an upside variance swap and the
centage change outside the corridor is subtracted from the square of the value d for a downside variance swap. The fixed payment F0c(L), c = u, d
total percentage change. is chosen at time t0 so that the corridor variance swap has zero initial cost
Our formulation in (1) treats entry and exit asymmetrically. Under our of entry. Suppose that Q 0c(L) is the known initial cost of creating the ter-
asymmetric formulation, there exists a model-free hedging strategy whose minal random payout Qnc(L), where c = u, d. Then the fair fixed payment
error is only third order, as shown in the next section. In contrast, suppose to initially charge on the corridor variance swap is simply given by:
that the exit payout was defined symmetrically to the entry payout to be
(ln L/Fi – 1)2. Then, there does not exist a model-free hedging strategy whose Q0c ( L)
F0c ( L) = , c = u, d (7)
error is only third order. B0 (tn )
A second reason for our asymmetric treatment of entry and exit is that
we want the sum of the payouts from an upside variance contract and a where B0(tn) denotes the initial price of a pure discount bond paying one
downside variance contract with the same barrier L to equal the payout dollar at tn. As a result, the next section focuses on determining an accu-
from a standard variance contract. A downside variance contract is defined rate approximation to Q 0c(L).
to be a financial security that has the following non-negative payout at the
fixed time tn: Approximate replication

Assumptions. For the rest of the article, we assume frictionless mar-
n   F 
2
 F
2 kets and deterministic interest rates. We also assume that one can trade fu-
Qnd ( L) ≡ ∑ 1Fi−1 ≤ L, Fi ≤ L  ln i  + 1Fi−1 > L, Fi ≤ L  ln i  tures at the same frequency (for example, daily) with which
i =1   Fi −1   L
 marking-to-market and swap monitoring occur. Finally, we assume that
one can take static positions in the continuum of European-style futures
 F  2  F  2   (2)
+1Fi−1 ≤ L, Fi > L  ln i  −  ln i    options with strikes inside the supporting corridor and maturing with the
 Fi −1   L   swap. Note that we make no assumptions regarding the stochastic process
 
followed by futures or option prices. In particular, jumps are allowed,
The payouts in (1) and (2) sum to the following payout of a standard vari- volatility can be stochastic, and the process parameters do not need to be
ance contract: known. Under the above assumptions, this section shows that the payouts
2
on upside and downside variance swaps can be well approximated by
n 
F  combining static positions in standard options with at most daily trading
Qn ( L) ≡ ∑  ln i  (3)
i =1  Fi −1 
in the underlying futures.
The market that probably best approaches the above idealised condi-
From (2), our treatment of entry and exit on the downside variance con- tions is that for S&P 500 derivatives, where one has liquid trading in both
tract is also asymmetric. In contrast, suppose that the exit payout for a futures and in European-style options. Although the S&P 500 index op-
downside variance contract was defined symmetrically with the entry pay- tions are written on the cash index, they often mature with the futures,
out to be (ln L/Fi – 1)2. Then the payouts to upside and downside variance and hence in those cases can be regarded as European-style futures op-
contracts with symmetrically defined exit and entry would not sum to the tions. Furthermore, as our replication error will be related to third and high-
payout (3) of a variance contract. The reason is that while the total return er moments of the underlying’s return, the reduction in these moments
decomposes into the returns to and from the barrier: arising from diversification in the index is attractive. We next review the
approximate replication of a variance swap payout before tackling the
Fi F L harder problem of approximately replicating the payout to a corridor vari-
ln = ln i + ln (4)
Fi −1 L Fi −1 ance swap.

Variance swap. It is well known that the geometric mean of a set of
the squared total return differs from the sum of squared returns to and positive numbers is never greater than the arithmetic mean. It is also well
from the barrier by twice the product of these returns: known that the larger the variation in the set of numbers, the greater the
2 2
disparity between the two means. The approximate replication of the pay-
2
 Fi   Fi   L   Fi   L  out to a variance swap exploits this basic property.
ln
 F  =  ln  +  ln  + 2  ln L  ln  F  (5)
i −1 L  Fi −1 i−1
By Taylor series expansion of ln F about F = Fi – 1, we note that:
3
Hence, if entry and exit are defined symmetrically for both upside and 1 1  ∆F 
downside variance contracts, the payout to a portfolio of an upside and
ln Fi − ln Fi −1 = ∆Fi − (∆Fi )2 + O  F i  , i = 1,..., n (8)
Fi −1 2 Fi 2−1  i −1 
downside variance contract would miss the payout to a variance contract
by the last term in (5). where ∆Fi ≡ Fi – Fi – 1 denotes the change in the futures price over day i.
The asymmetry of our payout definition in (1) vanishes if one assumes Rearranging implies that the squared daily return is just twice the differ-
continuous price processes, continuous path monitoring and the ability to ence between the daily compounded return and the continuously com-
trade the underlying continuously. Under these assumptions, the hedging pounded return, up to terms of order O(∆Fi/Fi – 1)3:
strategy we propose in the next section works perfectly. It should not be
too surprising that the relaxation of these idealised conditions necessarily 2 3
introduces replication error. What is perhaps surprising is that the replica-  ∆Fi   ∆Fi  Fi    ∆Fi 
 F  = 2  F − ln  F   + O  F  , i = 1,..., n (9)
tion error can be kept to third order provided one is willing to treat entry i −1  i −1 i −1 
 i −1
and exit asymmetrically.

XX RISK FEBRUARY 2004 ● WWW.RISK.NET

Squaring both sides of (8) implies: L
2
n F  n 
2 2 ⌠ 2
∑  ln F i  = ∑  F − L  ∆Fi + ⌡ 2 ( K − Fn ) dK
2 2 3 +
 Fi   ∆Fi   ∆Fi 
 ln F  =  F  + O  F  , i = 1,..., n (10) i =1 i −1 i =1 i −1 K
0
i −1 i −1 i −1
∞ 3
⌠ 2
n  ∆Fi 
2( n
F − K ) dK − u ( F0 ) + ∑ O 
+ (16)
Substituting (10) in (9) implies that the squared continuously compound- + 
ed return is just twice the difference between the daily compounded re- ⌡K F i =1  i −1
L
turn and the continuously compounded return, up to terms of order
O(∆Fi/Fi – 1)3: Thus, the payout to a variance swap is well approximated by summing
the payouts from a dynamic position in futures and a static position in op-
2 3 tions and bonds. For the dynamic component, (16) indicates that one holds
 Fi   ∆Fi  Fi    ∆Fi 
e–yin(tn – ti)(2/Fi – 1 – 2/L) futures contracts from day i – 1 to day i, where yin
 ln F  = 2  F − ln  F   + O  F  , i = 1,..., n (11)
i −1  i −1 i −1 
 i −1 is the continuously compounded yield on day ti to maturity tn. For the sta-
tic component, (16) indicates that one holds (2/K 2)dK puts at all strikes
Summing over i gives a decomposition of the sum of squared returns: below L, (2/K 2)dK calls at all strikes above L, and one shorts u(F0) bonds.
If the initial cost of the approximate hedge is financed by borrowing, then
the repayment at tn is:
2 3
n  F  n
2 n n  ∆F 
∑  ln F i  = ∑ F ∆Fi − 2∑ (ln Fi − ln Fi −1 ) + ∑ O  F i  L ∞
⌠ 2 P0 K , tn ( )
⌠ 2 C K , tn ( )
i =1 i −1 i =1 i −1 i =1 i =1 i −1
 2 dK +  2 0 dK − u F0 ( ) (17)
n
2 n  ∆F 
3 ⌡K B0 tn ( )
⌡K B0 tn ( )
=∑ ∆Fi − 2 ln Fn + 2 ln F0 + ∑ O  i  (12) 0 L

i =1 Fi −1 i =1  Fi −1 
where P0(K, tn) and C0(K, tn) respectively denote the initial prices of puts
and calls struck at K and maturing at tn. If there is no charge for the third-
due to telescoping. Thus, up to third-order terms, the sum of squared re- order approximation error, then (17) is the fair (non-annualised) fixed pay-
turns decomposes into the payout from a dynamic futures strategy and a ment for a variance swap on $1 of notional. This fixed payment is actually
function f(Fn) = –2lnFn + 2lnF0 of just the final futures price. As a static independent of the choice of L, since it only depends on the convexity of
position in bonds and options can be used to create this final payout func- the payout.
tion, approximate replication is feasible. One can interpret the dynamic component of our approximate repli-
There is some flexibility in choosing the composition of the replicating cating strategy as a Black (1976) model dynamic hedge to the static port-
portfolio since any linear function added to f can be offset by the appro- folio described above. By the Black model dynamic hedge, we have in
priate position in bonds and futures. As our ultimate goal is to approxi- mind that the hedger trades futures continuously under the belief that the
mately replicate the payout to a corridor variance swap, we will add a futures price process is continuous with constant volatility σ. As is well
linear function to f so that it becomes U-shaped with the minimum oc- known, the number of futures held at any time in this model is given by
curring at L. Hence, for any L > 0, suppose we subtract and add 2lnL + 2 the first partial derivative of the value function with respect to the futures
/L × (Fn – F0) to the right-hand side of (12): price. To show that the dynamic component of our hedge can be inter-
preted as a Black model dynamic hedge, let:
2 3
n  F  n
2 2 n  ∆F 
∑  ln F i  = ∑ F ∆Fi − L ( Fn − F0 ) + u ( Fn ) − u ( F0 ) + ∑ O  F i  (13)
L ∞
⌠ 2 ⌠ 2
U ( Fn ) ( ) ( )+ dK − u ( F0 )
+
i =1 i −1 i =1 i −1 i =1 i −1 ≡  2 K − Fn dK +  2 Fn − K
⌡K ⌡K
0 L
where:
 F − F0 F  (18)
= u ( Fn ) − u ( F0 ) = 2  n − ln n 
F − L F  L F0 
u (F ) ≡ 2  − ln  (14)
 L L
be the U-shaped payout created by the static position in bonds and op-
As shown in Carr & Madan (1998), any continuous payout at tn of just tions. Since u(L) = 0 by (14), U takes its minimum value of –u(F0) at Fn =
the final futures price can be spanned by the payouts from a static posi- L. The Black model value at time ti – 1 of this payout is given by:
tion in bonds and European-style options maturing at tn. To determine the
 F − F0 F 
( ) − yi −1,n (tn − ti −1 )
replicating portfolio for the payout u(Fn), note that the function u(F) is U-
V Fi −1,ti −1 ≡ e 2  i −1 − ln i −1  − σ 2 (tn − ti −1 )
shaped with zero value and slope at F = L. The second derivative u′′(F) = L F0 

2/F 2 > 0. Hence, a Taylor series expansion with second-order remainder
of u(Fn) about Fn = L implies: Hence, the Black model delta at time ti – 1 is:

∞ ∂  2 2
V ( Fi −1, ti −1 ) = e i−1,n ( n i−1 ) 
L −y t −t
⌠ 2 ⌠ 2 −  (19)
u ( Fn ) ( ) ( )
+ +
=  2 K − Fn dK +  2 Fn − K dK (15) ∂F  Fi −1 L 
⌡K ⌡K
0 L
which differs from the number of futures needed to hedge a variance swap
Now: only by a small present value factor. Surprisingly, the Black model delta
in (19) is actually independent of σ, that is, ∂2V/∂σ∂F = 0. Put another way,
n
Fn − F0 = ∑ ∆Fi the static portfolio is chosen so that its Black model vega is independent
i =1 of the futures price. Of course, the Black model dynamic hedge of this
portfolio only works perfectly under continuous trading, continuous price
and substituting this and (15) into (13) implies: paths and constant volatility. Since the approximate replicating strategy ac-

WWW.RISK.NET ● FEBRUARY 2004 RISK XX

Cutting edge l Strap?

1. Function ur with L = 1 variance contracts is just the static option position in the hedge for the
variance contract.
For the dynamic components of the proposed hedges, recall that we
0.7 delta-hedge each option at σ = 0. Hence, for the dynamic component of
the proposed hedge for an upside variance contract, one holds –e–yin(tn –
0.6 ti)(2/L – 2/F +
i – 1) futures contracts from day i – 1 to day i, since out-of-the-
money call deltas vanish under zero volatility. For the dynamic compo-
0.5
nent of the proposed hedge for a downside variance contract, one holds
–e–yin(tn – ti)(2/Fi – 1 – 2/L)+ futures contracts from day i – 1 to day i, since
0.4
Payout

out-of-the-money put deltas also vanish under zero volatility. We again

0.3 note that the sum of the dynamic futures positions in the proposed hedges
for the upside and downside variance contracts is just the dynamic futures
0.2 position in the hedge for the variance contract.
If the futures price opens and closes below L, then in the proposed
0.1 hedge for the upside variance contract, no futures are held and there is no
gain in the bonds or out-of-the-money calls marked at zero volatility. Like-
0 wise, the intrinsic value of the upside variance contract does not change
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 in this case, so the proposed hedge works perfectly in this case. Further-
Futures price
more, the proposed hedge for a downside variance contract must also work
in this case since this hedge is just the difference in the successful hedges
of a variance contract and an upside variance contract.
tually involves only discrete trading at prices that can reflect jumps and If the futures price opens below L and closes above, then no futures
stochastic volatility, one would anticipate that this attempt at a Black model are held in the proposed hedge for the upside variance contract, and the
hedge would fail. However, for the particular static portfolio described intrinsic value of the call portfolio rises from zero to ur(Fi). Figure 1 shows
above, the hedging error approximates the payout to a variance swap with ur as a function of F when L = 1. The first derivative is u′r(F) = (2/L – 2/F)+,
fixed payment σ2(tn – t0). which is continuous at F = L. The second derivative is u′′r (F) = 1F > L 2/F2,

Upside and downside variance swaps. The last subsection showed which is discontinuous at F = L. By a Taylor series expansion of u(Fi) about
that the payout to a variance swap: Fi = L:
2 3
n  F   ∆F 
2
 ∆F 
∑  ln F i  − σ (t n − t 0 ) ur ( Fi ) =  i  + O  i 
2
(22)
i =1 i −1   L   Fi −1 

could be approximately replicated by forming a static portfolio of options for Fi > L since |Fi – L| ≤ |∆Fi|. On days when the futures price enters the
and bonds that has the U-shaped payout U(Fn). This portfolio is delta- upper corridor, the intrinsic value of the upside variance contract rises by
hedged daily with the Black model delta for each option calculated using (ln Fi/L)2. From (10), this differs from ur(Fi) by O(∆Fi /Fi – 1)3, so the pro-
the fixed volatility rate σ. It follows that if we just wish to create the pay- posed hedge is sufficiently accurate in this case as well. Furthermore, the
out to a variance contract: proposed hedge for a downside variance contract must also work in this
2
second case for the same reason as in the first case.
n  F  If the futures price opens and closes above L, then the analysis of the
∑  ln F i 
i −1 
i =1 last subsection implies that the proposed hedge of the upside variance con-
tract has a profit and loss of (∆Fi/Fi – 1)2 + O(∆Fi/Fi – 1)3, while the intrin-
we could delta-hedge each option at σ = 0. sic value of the upside variance contract rises by (ln Fi/Fi – 1)2. From (10),
To approximate the payouts to upside and downside variance contracts, the proposed hedge has sufficient accuracy in this third case. Furthermore,
suppose we guess that the approximate hedge just involves delta-hedging the proposed hedge for a downside variance contract must also work in
options struck above and below the barrier respectively, where each op- this case.
tion is delta-hedged at zero volatility. Hence, for the upside variance con- If the futures price opens above L but closes below it, then the analy-
tract, the static component of the proposed hedge has a payout that is sis of the profit and loss from the proposed hedge of the upside variance
constant for Fn ≤ L and is given by the right half of the U-shaped payout contract is complicated by the fact that ur(F) defined in (20) is not an an-
U(Fn) defined in (18) for Fn > L: alytic function of F. However, we note that exit of the upper corridor is
equivalent to entry of the lower corridor. If the futures price enters the
U r ( Fn ) ≡ ur ( Fn ) − ur ( F0 ) , where ur ( F ) ≡ u ( F )1F > L (20) lower corridor from above, then no futures are held in the proposed hedge
to the downside variance contract and the intrinsic value of the puts held
To create the payout Ur(Fn), we would only hold the (2/K 2)dK calls at all rises from zero to:
strikes K above the lower bound L and we would also short ur(F0) bonds. 2 3
 ∆F   ∆F 
No puts would be held. u
( Fi ) =  i  + O  i  , Fi < L (23)
Similarly, the proposed hedge for the downside variance contract has  L   Fi −1 
a static component payout that is constant for Fn ≥ L and given by the left
half of the U-shaped payout U(Fn) defined in (18) for Fn < L: from a Taylor series expansion of u(Fi) about Fi = L. When the futures
price enters the lower corridor, the intrinsic value of the downside vari-
U
( Fn ) ≡ u
( Fn ) − u
( F0 ) , where u
( F ) ≡ u ( F )1F < L (21) ance contract rises by (ln Fi /L)2, which only differs from u
(Fi) by O (∆Fi /Fi
3
– 1) . From (10), the proposed hedge to the downside variance contract has
Hence, we hold (2/K 2)dK puts at all strikes K below L and we short sufficient accuracy in this last case. It follows that the proposed hedge for
u
(F0) bonds. No calls are held. We note that the sum of the static po- the upside variance contract must also work in this case since this hedge
sitions in options in the proposed hedges for the upside and downside is just the difference in the successful hedges of a variance contract and

XX RISK FEBRUARY 2004 ● WWW.RISK.NET

the downside variance contract. 2. Distribution of profit and loss under
We have just shown that the payout Qun(L) of the upside variance con- geometric Brownian motion
tract is well approximated in all cases:
∞
⌠ 2 50
Qnu ( L) = 
⌡L K 2
( Fn − K )+ dK − ur ( F0 )
Asymmetric
+ 3 Symmetric
n 
2 2  n  ∆F  (24) 40
− ∑ −  ∆Fi + ∑ O  i 
i =1  L Fi −1  i =1  Fi −1 
30
If the initial cost of the approximate hedge is financed by borrowing, then
the repayment at tn is:
20
∞
⌠ 2 C0 K , tn ( )
 dK − ur F0 ( ) (25)
⌡L K 2 B0 tn ( ) 10

If there is no charge for the third-order approximation error, then this is

the fair (non-annualised) fixed payment for an upside variance swap on 0
$1 of notional. The corresponding entity for a downside variance contract –0.05 –0.03 –0.01 0.01 0.03
is:
L
⌠ 2 P0 K , tn ( )
 dK − u
F0 ( ) (26) 3. Distribution of profit and loss under
⌡0 K 2 B0 tn ( )
jump diffusion
Monte Carlo simulation
This subsection reports the distribution of hedging errors arising from sim- 50
ulating the hedge of an upside variance swap over 250,000 paths. We con- Asymmetric
sidered the errors arising from hedging the sale of an upside variance swap Symmetric
40
with daily monitoring, a three-month term, and a lower barrier of $90. We
first assumed that the underlying futures price starts at $100 and follows
geometric Brownian motion with 5% real world drift and 30% volatility. 30
Figure 2 shows the density function of the hedging errors. The units on
the x-axis correspond to the fixed leg price of 900 = 10,000(30%)2. The
20
curve designated symmetric corresponds to an upside variance swap pay-
out where exit and entry are treated symmetrically. The curve designated
asymmetric uses the asymmetric upside variance swap payout proposed 10
in this article. The asymmetric profit and loss has mean –0.0097 and stan-
dard deviation 0.01 while the symmetric profit and loss has mean –0.042
and standard deviation 0.069. Note that the symmetric profit and loss has 0
–0.05 –0.03 –0.01 0.01 0.03
a fat tail for losses.
We next assumed that the underlying futures price follows the jump
diffusion process suggested by Merton (1976). The initial futures price
start at $100 and has 5% real world drift, and a 30% diffusion coefficient and Xi ≡ H if Fi > H. The payout on the corridor variance contract is now
as before. We set the arrival rate equal to one, so that jumps arrive once defined as:
a year on average. The jump in the log price is normally distributed with
n 
2 2
mean zero and standard deviation of 10%. Figure 3 summarises the re-   F   F 
sults of the previous simulation but now using the Merton model. The Qn ( L, H ) ≡ ∑ 1Fi−1 ∈C , Fi ∈C  ln i  + 1Fi−1 ∉C , Fi ∈C  ln i 
i =1   Fi −1   N i −1 
hedges corresponding to the asymmetric payout definition continue to 
perform well, while the same hedges coupled to a symmetric payout de-  F  2  F  2 
finition are actually ‘short jumps’. The asymmetric mean is –0.0037 with + 1Fi−1 ∈C , Fi ∉C  ln i  −  ln i  
 Fi −1   Xi  
standard deviation 0.094 while the symmetric mean is –0.08 with standard  
deviation 0.3. 2 2  (27)
 F − L F −H
These simulation results clearly suggest that the complications arising + 1Fi−1 < L, Fi > H + 1Fi−1 > H , Fi < L   i − i 
 L   H 
from an asymmetric payout definition are outweighed by the improved 
hedge effectiveness.
The first three terms in the summand correspond to the three terms in the
Interior corridor summand in (1). The last term arises from the possibility of jumping over
In this section, we generalise our results on upside and downside variance the corridor in either direction. In either case, the squared return from the
swaps to corridor variance swaps with an interior corridor. Let C denote exit price is subtracted from the squared return from the entry price.
the interior corridor (L, H) where L > 0 and H is finite. The futures price To create the payout in (27), consider the following function:
is now said to enter the corridor on day i if Fi – 1 ∉ C and Fi ∉ C. In this
case, define the entry price Ni – 1 as Ni – 1 ≡ L if Fi – 1 < L and Ni – 1 ≡ H if 1F > L u ( F ) if F ≤ H
φ(F ) ≡  H
Fi – 1 > H. The futures price is now said to exit the corridor on day i if Fi 2  − 1 − ln
  L ( ) + (
H
L
2
L
− 2
H )( F − H ) if F > H
(28)

– 1 ∈ C and Fi ∉ C. In this case, define the exit price Xi as Xi ≡ L if Fi < L

WWW.RISK.NET ● FEBRUARY 2004 RISK XX

Cutting edge l Strap?

4. Function φ with L = 1 and H = 2 symmetrically. As a result, the sum of a downside variance swap and an
upside variance swap is a standard variance swap, which does not remain
true when exit and entry are treated symmetrically.
1.4 There are at least seven extensions to this work. First, one can further
analyse our small approximation error to see if it can be at least partially
1.2 spanned. For example, a simple linear regression of the error on a con-
stant and the change in the futures price can be used as a guide to how
1.0 to account for this error in determining the fixed rate for the corridor vari-
ance swap and the dynamic component of the hedge. Using ordinary least
0.8
Payout

squares, one finds that the return skewness affects the fixed rate, while
0.6 the return kurtosis affects the futures position. Second, one can try to relax
our model assumptions such as continuum of strikes, deterministic inter-
0.4 est rates and frictionless futures trading, or at least try to determine their
effect. Third, one can attempt to determine the effect of small perturba-
0.2 tions in our definitions. For example, (10) makes it clear that the hedge
has the same order error if daily returns are discretely compounded rather
0 than continuously compounded. One can also try to determine the effect
0.5 1.0 1.5 2.0 2.5 of demeaning the daily returns as some (corridored) variance swaps have
Futures price
this feature. Fourth, one can adapt our approach to make it more applic-
able to the corridor variance realised from individual stock returns. If
stocks replace single-name futures as the underlying, then stock dividends
where u(F) is defined in (14). Thus φ (F) = ur(F) for F ≤ H and is the tan- become relevant. Also, if listed options are used in the hedge, then Amer-
gent to ur at F = H for F > H. Figure 4 graphs φ against F. The function φ ican-style options must be handled. Fifth, one can supplement the ap-
is continuous and differentiable everywhere, but it is not twice differen- proximation developed here by also developing bounds on the fair fixed
tiable at L and at H. payment via super- and sub-replication of the corridor variance. Sixth, it
We can use calls maturing at tn to create the payout φ(Fn): would be interesting to extend this work by characterising the entire class
of path-dependent payouts that can be approximated or bounded in this
H
⌠ 2 way. Finally, one can also try to develop a theory of model-free approx-
φ ( Fn ) =  ( Fn − K )+ dK (29)
⌡L K 2 imate hedging that would in general allow semi-dynamic trading in both
futures and options. In the interests of brevity, these extensions are best
Using an analysis similar to that in the last section, we conclude that: left for future research. ■
H
⌠ 2 Peter Carr is head of quantitative research at Bloomberg and director of
Qn ( L, H ) = 
⌡L K 2
( Fn − K )+ dK − φ ( F0 ) the Masters in Mathematical Finance Program at the Courant Institute,
+ 3 New York University. Keith Lewis is an independent consultant. The
n   n  ∆F  (30)
2 2
− ∑ −  ∆Fi + ∑ O  i  views expressed herein represent only those of the authors and do not
i =1  L Fi −1 ∧ H  i =1  Fi −1  necessarily reflect the views of their employer or clients. The authors
thank two anonymous referees, Dean Curnutt, Zhenyu Duanmu, Jim
Thus, the desired payout is again well approximated by the sum of the Gatheral, Dilip Madan, Reiner Martin, Alex Mayus, Ragu Raghavan,
payout from a static position in calls and bonds with the payouts from a Satish Ramakrishna and Pav Sethi for their perspectives. They are not
dynamic position in futures. For the static component, one holds (2/K2)dK responsible for any errors. Email: [email protected], [email protected]
calls at all strikes in the corridor (L, H). One also borrows φ(F0) pure dis-
count bonds paying $1 at tn. For the dynamic component, one holds –e–yin(tn
REFERENCES
i – 1∧H) futures contracts from day i – 1 to day i. When Fi – 1
– ti)(2/L – 2/F +

≤ L, no futures are held, as in the last section. When Fi – 1 > H, the num-
Black F, 1976
ber of futures contracts held is independent of Fi – 1, so that the number The pricing of commodity contracts
of contracts held hardly changes while the underlying remains above the Journal of Financial Economics 3, pages 167–179
corridor. If the initial cost of the approximate hedge is financed by bor-
rowing, then the repayment at tn is: Carr P and D Madan, 1998
Towards a theory of volatility trading
In Volatility, edited by R Jarrow, Risk Publications, pages 417–427. Reprinted in
H
⌠ 2 C0 K , tn ( )
 dK − φ F0 ( ) Option Pricing, Interest Rates, and Risk Management, edited by Musiella, Jouini
⌡L K 2 B0 tn ( ) (31) and Cvitanic, Cambridge University Press, 2001, pages 458–476. Available at
www.math.nyu.edu/research/carrp/papers

If there is no charge for the third-order approximation error, then this is Chriss N and W Morokoff, 1999
the fair (non-annualised) fixed payment for the interior corridor variance Market risk for volatility and variance swaps
Risk October, pages 55–59
swap on $1 of notional.
Demeterfi K, E Derman, M Kamal and J Zhou, 1999
Summary and extensions A guide to variance swaps
We defined the payout to a corridor variance swap in such a way that the Risk June, pages 54–59
payout could be well approximated by the payout from combining static
Merton R, 1976
positions in options and bonds with at most daily trading in the underly- Option pricing when underlying stock returns are discontinuous
ing futures. Although our payout definition treats entry and exit asym- Journal of Financial Economics 3, pages 125–144
metrically, it treats entry for a downside variance swap symmetrically with
entry for an upside variance swap. Exit for the two swaps is also treated

XX RISK FEBRUARY 2004 ● WWW.RISK.NET