Alexander Gairat

In collaboration with IVolatility.com

Variance swaps

Introduction
The goal of this paper is to make a reader more familiar with pricing and hedging variance swaps and to propose some practical
recommendations for quoting variance swaps (see section Conclusions).

We give basic ideas of variance swap pricing and hedging (for detailed discussion see [1]) and apply this analyze to real
market data. In the last section we discuss the connection with volatility swaps.

A variance swap is a forward contract on annualized variance, the square of the realized volatility.

i i S@ti+1 D yy2
„ j
j j
j z
zz
z
n−1
1
k k S@ti D {{
VR = σR = 252
2
Log
n−2
i=1

Its payoff at expiration is equal to

PayOff = σR 2 − KR 2

Where

S@ti D - the closing level of stock on ti valuation date.

n - number of business days from the Trade Date up to and including the Maturity Date.

When entering the swap the strike KR is typically set at a level so that the counterparties do not have to exchange cash flows
(‘fair strike’).

Variance swap and option delta-hedging.

Advantage of trading variance swap rather than buying options is that it is pure play on realized volatility no path dependency is
involved. Let us recall how path dependency appears in trading P&L of a delta-hedged option position.

If we used implied volatility σi on day i for hedging option then P&L at maturity is sum of daily variance spread weighted by
dollar gamma.

‚ Hri 2 − σi 2 ∆tL Γi Si 2
1 n−1
Final P & L =
2 i=1

Si+1 −Si
ri = is stock return on day i

Γi Si 2 = Γ@i, Si D Si 2 - dollar gamma on day i.

Si
2 Variance Swaps

So periods when dollar gamma is high are dominating in P&L.

Market example
We use here historical data for the option with one year expiration on SPX500, strike K=1150, observation period from
2004/1/2 to 2004/12/4

5 days average of ri 2 −σi 2 ∆t

0.00005

4ê1ê2 4ê3ê1 4ê4ê26 4ê6ê22 4ê8ê17 4ê10ê12 4ê12ê7

-0.00005

-0.0001

5 days average of dollar gamma Γt St 2

10000
8000
6000
4000
2000
4ê1ê2 4ê3ê1 4ê4ê26 4ê6ê22 4ê8ê17 4ê10ê12 4ê12ê7
5 days average of plain variance spread and variance spread weighted by dollar gamma

4ê1ê2 4ê3ê1 4ê4ê26 4ê6ê22 4ê8ê17 4ê10ê12 4ê12ê7

-0.002

-0.004

-0.006

-0.008

A∗‚n−1
i=1 Hri −σi ∆tLΓi Si
2 2 2

‚n−1
i=1 Hri −σi ∆tL
2 2

Where dollar gamma is normalized by factor A = 252 ê H⁄ Γi Si 2 L

Variance Swaps 3

Pricing
The fair strike KR can be calculated directly from option prices (under assumptions that the underlying follows a continuous
diffusion process, see Appendix 1)

„ Put@Ki D + „ Call@Ki D
2 ∆Ki 2 ∆Ki
(1) KR 2 = rT rT
T 2 T 2
Ki ≤FT Ki Ki >FT Ki

where

FT = Hr−qL T is the forward price on the stock at expiration time T,

r risk-free interest rate to expiration,

q dividend yield.

Example: VIX
VIX CBOE volatility index it is actually fair strike KR for variance swap on S&P500 index.

1 i FT y
„ Put@Ki D + „ Call@Ki D − j
j − 1z
z
2 ∆Ki 2 ∆Ki 2

T k K0 {
VIX@TD2 = rT rT
T Ki ≤FT Ki 2 T Ki >FT Ki 2

where K0 is the first strike below the forward index level FT .

The only difference with formula (1) is the correction term

1 i FT y
j
j − 1z
z
2

T k K0 {
−

which improves the accuracy of approximation (1).

There two problems in calculation the fair strike of variance swap in from raw market prices:

è We need quotes of European options.

They are available for indexes but not for stocks.

è Number of available option strikes can be not sufficient for accurate calculation.

The first problem can be solved by calculating prices of European options from implied volatilities of listed American options.
To overcome the second problem we could use additional strikes and interpolate implied volatilities to this set.
4 Variance Swaps

Examples
In this section we calculate prices of variance swap which starts on n-th trading day counted from 1.1.2004 with the expiration
on 31.12.2004. We compare two values:

The first KR is calculated based on 40 European options with standardized moneyness x

è!!!!
x = Log@K ê FT D ë Iσ tM

x in range from -2 to 2 with step 0.1.

The second value KR HRawL is calculated based on the raw market prices (either they are European or American). We use only
the strikes of options with valid implied volatilities IV, i.e. IV is in range from 0 to 1. These strikes we call valid strikes. The
number of valid strikes we denote by N[K].

SPX
The differences here are quite small less than 1 volatility point. It is natural to expect small difference for SPX because in both
cases we used European option and, what is more important, the number of valid strikes is rather large more than 40.

25

22.5

20

17.5

50 100 150 200

12.5

KR KR HRawL
Variance Swaps 5

KR −KR HRawL

50 100 150 200

-0.2

-0.4

-0.6

-0.8

N@KD
50
45
40

50 100 150 200

EBAY INC

gence happens due to the small number of strikes available for calculation of KR HRawL. Still the number of valid strikes is
This example is different. Only American options are available. But still deference are not too large. And we see that diver-

relatively large close to 20.

42.5

40

37.5

35

32.5

30

27.5

50 100 150 200 250

KR KR HRawL
6 Variance Swaps

KR −KR HRawL

2.5

1.5

0.5

50 100 150 200 250

-0.5

-1
N@KD
20
18
16
14
12
10
50 100 150 200 250

In the following two examples we will see that the quality of approximation (1) with raw market prices drops dramatically due
to small number of valid strikes.

ADVANCED MICRO DEVICES

70

65

60

55

50

45

50 100 150 200 250

KR KR HRawL
Variance Swaps 7

KR −KR HRawL

15

10

50 100 150 200 250

-5

-10

-15

-20
N@KD
14
12
10
8

50 100 150 200 250

Time Warner INC

30

25

20

50 100 150 200 250

KR KR HRawL
8 Variance Swaps

KR −KR HRawL

10

50 100 150 200 250

-5

N@KD
10
8
6
4

50 100 150 200 250

Hedging
Replication strategy for variance V[T] follows from the relation (for details see Appendix)

i i S@ti+1 D yy in−1 S@ti+1 D − S@ti D

2 j y
z
„j
jLog j
j z
zz
z ≈ j
j − Log@ST ê S0 Dz
z
j‚
n−1

z
2
1 1
k k S@ti D {{ T ki=1 S@ti D {
V@TD ≈
T T
i=1

The first term in the brackets

S@ti+1 D − S@ti D
‚
n−1

i=1 S@ti D

can be thought as P&L of continuous rebalancing a stock position so that it is always long 1 ê St shares of the stock.

The second term

−Log@ST ê S0 D

represent static short position in a contract which pays the logarithm of the total return.

Example
As an example we take the stock prices of PFIZER INC, for one year period from 1/1/2004 to 1/1/2005.
Variance Swaps 9

St PFIZER INC

38

36

34

32

30

28

26

Trading days from 1ê1ê2004

50 100 150 200 250

For this stock prices we calculate realized variance Vt and its replication Πt for each trading day

Πt = 2 ‚ HS@ti+1 D − S@ti DL ê S@ti D − 2 Log@St ê S0 D

ti <t

Difference of realized variance and replication in volatility points

è!!!!!!!!!!!! è!!!!!!!!!!!!
100H Vt ê t − Πt ê t L

0.1

0.05

50 100 150 200 250

-0.05

-0.1

The typical differences in this plot are less than 0.1 volatility point. Large differences in the first 10 days are explained by
large ratio of expiration period to time step - 1 day. Large difference around 240-th trading day appears due to the jump of the
stock price.
10 Variance Swaps

We see that combination of dynamic trading on stock and log contract can efficiently replicate variance swap.

The payoff of log contract can be replicated by linear combination of puts and calls payoffs (see Appendix)

−Log@ST ê S0 D ≈ − +„ HKi − ST L+ + „ HST − Ki L+

ST − S0 ∆Ki ∆Ki
S0 Ki ≤S0 Ki 2 Ki >S0 Ki 2

Hence log contract can be replicated by standard market instruments:

è short position in 1/S forward contracts strike at S0

è long position in ∆Ki ê Ki 2 put options strike at K, for all strikes Ki from 0 to S0 ,

è long position in ∆Ki ê Ki 2 call options strike at K, for all strikes Ki > S0

So this portfolio (with doubled positions) and dynamic trading on stock replicates variance swap. The replication also gives the
fair strike value of volatility swap at time t

2 i
j
j S@ti+1 D − S@ti D y
z
KR @tD2 ≈ j
j
j ‚ + Log@FT @tD ê St Dz
z
z
z
T kti+1 <t S@ti D {
(2)

Hdynamic tradingL
2 i j S0 − FT @tD + „
j
j
∆Ki r HT−tL
T k
+ Put@St , Ki , T − tD +
S0 2

y
Ki ≤S0 Ki

„ Call@St , Ki , T − tDz z Hstatic replicationL

z
∆Ki r HT−tL
{
2
Ki >S0 Ki

where r is interest rate.

Market examples

è!!!!!!!!!!!!!!!!!!!!
In this section we test replication performance on a set of market data.

For each day we calculate price of volatility swap Vt @TD ê T started on 1.1.2004 with expiration on 1.1.2005 and the
price of replication strategy KR @tD.

At time t the volatility swap consists of the realized variance and the expected future variance

i i S@ti+1 D yy2
Vt @TD = „ j
j j
j z
zzz +
1 1
k k S@ti D {{
Log
T T <t
´¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨ ≠¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨Æ
t i+1
realized variance

i
j y
r HT−tL j
j„ Put@St , Ki , T − tD + „ Call@St , Ki , T − tDz z
z
2 ∆Ki ∆Ki
k {Æ
T 2 2
Ki ≤FT Ki Ki >FT Ki
´¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨
¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨
¨¨¨¨¨¨¨¨≠¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨
¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨
¨¨¨¨¨¨¨
expected variance
Variance Swaps 11

For replication we use 40 European options with strikes K such that the standardized moneyness x

è!!!!
x = Log@K ê FT @0DD ë Iσ TM

x is in n range from -2 to 2 with step 0.1.

option prices and get the fair strike value KR @tD from Eq.(2).
The implied volatilities for these strikes we get by linear interpolation of the market implied volatilities. Then we calculate

To estimate the performance of replication we define the replication error

i
j y
z
j$%%%%%%%%
1 %%%%%%%%%%% z
Error@tD = 100 ∗ j
j
j Vt @TD − KR @tDz
z
z
j z
k {
T

We also plot the number of valid market strikes N[K] (strikes with well defined implied volatilities).
12 Variance Swaps

Replication Error

SPX

KR

18

16

14

50 100 150 200

Error
0.4
0.3
0.2
0.1

-0.1 50 100 150 200

-0.2

N@KD
50
45

40

50 100 150 200

Variance Swaps 13

PFIZER INC

KR

26

25

24

23

22

21

50 100 150 200 250

Error
0.2

50 100 150 200 250

-0.2
-0.4
-0.6

N@KD

14
12
10
8
6

50 100 150 200 250

14 Variance Swaps

EBAY INC

KR
38

37

36

35

34

50 100 150 200 250

Error

-0.25 50 100 150 200 250

-0.5
-0.75
-1
-1.25
-1.5

N@KD
20
18
16
14
12
10
50 100 150 200 250
Variance Swaps 15

ADVANCED MICRO DEVICES

KR

60

57.5

55

52.5

50 100 150 200 250

47.5

Error
0.5

50 100 150 200 250

-0.5
-1
-1.5

N@KD
14

12

10

50 100 150 200 250

16 Variance Swaps

GENERAL MOTORS CORP

KR

32

30

28

26

24

50 100 150 200 250

Error

0.6

0.4

0.2

50 100 150 200 250

N@KD
12

10

50 100 150 200 250

Variance Swaps 17

Time Warner Inc.

KR
30

28

26

24

22

50 100 150 200 250

Error
0.6
0.5
0.4
0.3
0.2
0.1

-0.1 50 100 150 200 250

N@KD
10

50 100 150 200 250

18 Variance Swaps

Citigroup Inc

KR

24

22

20

18

50 100 150 200 250

Error
0.5
0.4
0.3
0.2
0.1

-0.1 50 100 150 200 250

-0.2

N@KD
10
9
8
7
6
5
4

50 100 150 200 250

Variance Swaps 19

EXXON MOBIL CORP

KR

21

20

19

18

50 100 150 200 250

Error

50 100 150 200 250

-0.1
-0.2
-0.3
-0.4
-0.5

N@KD

12
10
8
6
4

50 100 150 200 250

20 Variance Swaps

MORGAN STANLEY DEAN WITTER CORP

KR

32

30

28

26

50 100 150 200 250

Error
0.2

50 100 150 200 250

-0.2

-0.4

N@KD

10

50 100 150 200 250

Variance Swaps 21

FORD MOTOR CORP

KR

40

38

36

34

32

30

50 100 150 200 250

Error

0.1

50 100 150 200 250

-0.1
-0.2
-0.3

N@KD
10
8
6
4

50 100 150 200 250

22 Variance Swaps

GUIDANT CORP

KR

34

32

30

50 100 150 200 250

Error
0.2
0.1

50 100 150 200 250

-0.1
-0.2
-0.3

N@KD
12
10
8
6
4

50 100 150 200 250

Variance Swaps 23

TENET HEALTHCARE CORP

KR
70

65

60

55

50

50 100 150 200 250

Error

3
2.5
2
1.5
1
0.5

-0.5 50 100 150 200 250

N@KD

12
10
8
6
4

50 100 150 200 250

24 Variance Swaps

MICRON TECHNOLOGY INC

KR

50
48
46
44
42
40
38

50 100 150 200 250

Error
0.5
0.25

-0.25 50 100 150 200 250

-0.5
-0.75
-1

N@KD
14
12
10
8
6
4

50 100 150 200 250

Variance Swaps 25

Conclusion
In section Pricing we tested two methods for estimation variance swap price. One method is based on the raw option prices.
Another is based on a virtual set of European options with implied volatilities extracted from the market data. We saw that the
both methods give very close results in case of large number of valid market strikes. But if this number is small (less than
15-10) the second method gives more regular historical prices. Under condition that implied volatilities are available this
method is simple, fast and can be recommended for quoting variance swaps.

In section Hedging we tested replication strategy for hedging variance swaps. The results confirm that described replication is
quite robust, the error of replication is typically less than one volatility point. However practical attractiveness of this method
can be restricted by small number of listed options on underlying. In this case trader needs to use OTC market to construct
replication portfolio.
26 Variance Swaps

Variance and Volatility swaps

It is interesting to compare fair strike level KR for variance swap with ATM Implied Volatility (IVATM ). This topic is exten-
sively discussed in [2] (see also [3] ) with examples of VIX and VXO. This comparison is interesting if we think of IVATM as
an estimate of future volatility σR . In this case it is natural to expect that there is a positive spread between KR and IVATM . It
follows from convexity adjustment argument (see Appendix). Let us recall it.
¯
Suppose the future variance V@TD has mean V and variance W (under risk-neutral measure)
¯
V = E V@TD,
W = E HV@TD − WL.

Then
è!!!!!!!!!!!!!! "####
¯
KR = EV@TD = V

And
W
KR − σR ≈
8 KR 3

We plot here 5 days moving average of fair strike level KR calculated for the period (t,T) calculated by Eq. (1) and ATM
implied volatility of the same period. Indeed we see quite stable positive spread between KR and ATM implied volatility. If
interpretation of ATM implied volatility as expected future volatility really holds then from this spread we can estimate new
parameter W variance of variance. This parameter can be used for price estimates of volatility or variance derivatives.

KR , σR and KR − σR

SPX
20

15 KR

10 IVATM

5 KR −IVATM

50 100 150 200

Variance Swaps 27

PFIZER INC

30
25
KR
20
15 IVATM
10
KR −IVATM
5

50 100 150 200 250

EBAY INC
40

30 KR

20 IVATM

10 KR −IVATM

50 100 150 200 250

ADVANCED MICRO DEVICES

60
50
40 KR

30 IVATM
20
KR −IVATM
10

50 100 150 200 250

GENERAL MOTORS CORP

35
30
25 KR
20
15 IVATM

10
KR −IVATM
5

50 100 150 200 250

28 Variance Swaps

Time Warner INC

30
25
20 KR

15 IVATM
10
KR −IVATM
5

50 100 150 200 250

Citigroup INC
25

20
KR
15
IVATM
10

5 KR −IVATM

50 100 150 200 250

EXXON MOBIL CORP

20

15 KR

10 IVATM

5 KR −IVATM

50 100 150 200 250

MORGAN STANLEY DEAN WITTER

30
25
KR
20
15 IVATM
10
KR −IVATM
5

50 100 150 200 250

Variance Swaps 29

FORD MOTOR
40

30
KR

20 IVATM

10 KR −IVATM

50 100 150 200 250

GUIDANT CORP
35
30
25 KR
20
IVATM
15
10 KR −IVATM
5

50 100 150 200 250

TENET HEALTHCARE CORP

70
60
50 KR
40
30 IVATM
20
KR −IVATM
10

50 100 150 200 250

MICRON TECHNOLOGY INC

50

40
KR
30
IVATM
20

10 KR −IVATM

50 100 150 200 250

30 Variance Swaps

References
[1] K. Demeterfi, E. Derman, M. Kamal, J. Zou More Than You Ever Wanted To Know About Volatility Swaps. Quantita-
tive Strategies: Research Notes, Goldman Sachs 1999

[2] P.Carr and L.Wu A Tale of Two Indices http://www.math.nyu.edu/research/carrp/papers/pdf/vixov_florida3.pdf

[3] P.Carr and R. Lee Robust replication of volatility derivatives http://www.math.nyu.edu/research/carrp/papers

Appendix
Variance Replication
If stock price is follows continuous diffusion with volatility σt
St
= µt t + σt Wt
St

Then by Ito's lemma

HLog@St DL = µt
1
t + σt Wt − σt 2 t
2

Or

− Log@St D
1 St
σt 2 t=
2 St

Hence for total variance from 0 to T we get

V@TD = ‡ σt 2 t=2‡ − 2 Log@ST ê S0 D

T T St
(A1.1)
0 0 St

Log payoff decomposition

It is easy to check the following identity for any S>0

Log@ST ê SD = HST − SL − ‡ HK − ST L+ HST − KL+

∞

K−‡
S
1 1 1
(A1.2) K
S K2 K2
0 S

The fair strike value of variance swap van be calculated by taking expectation of future variance under risk-neutral measure at
time t

2 i
j y
KR @tD2 = jE ‡ − E Log@ST ê S0 Dz
z=
1 T Sτ
T k 0 Sτ {
E V@TD =
T
2 i
j y 2
j
j‡ + Hr − qL HT − tLz
z
z− E Log@ST ê S0 D
t Sτ
T k 0 Sτ { T
Variance Swaps 31

Using (A1.2) with S = S0

E Log@ST ê S0 D =

HFT @tD − S0 L − ‡
∞
r HT−tL
Put@K, T − tD K − ‡ r HT−tL
S0
1 1 1
Put@K, T − tD K
S0 K2 K2
0 S0

where

FT @tD = Hr−qL HT−tL

.

Finally we have
2 i
j y
KR @tD2 = j
j‡ + Hr − qL HT − tLz
z
z+
t Sτ
T k 0 Sτ {

2 i
j
j
j
y
z
z
j HFT @tD − S0 L + ‡ Put@K, T − tD Kz
z
∞

j Put@K, T − tD K + ‡ z
r HT−tL r HT−tL
S0

j z
z
1 1 1
T j S0
k {
K2 K2
0 S0

Pricing at initial moment, t=0

Expression for the fair strike KR = KR @0D at time-0 can be simplified in the following way
2 i
j y 2
jE ‡ − E Log@ST ê S0 Dz
z= HHr − qL T − E Log@ST ê S0 DL
1 T S

T k 0 St { T
t
KR 2 = E V@TD =
T

E Log@ST ê FT D
2
=−
T

Using (A1.2) with S = FT

−E Log@ST ê FT D = ‡
∞

Put@KD K + ‡
FT
1 rT
1 rT
Call@KD K
K2 K2
0 FT

Hence
i FT 1
j y
z
j
j z
j Call@KD Kz
z
∞

j
j‡ K2 Put@KD K + ‡ K2 z
z
rT

j z
1
KR =
2 rT

k0 {
T
FT

Convexity Adjustment
32 Variance Swaps

KR

convexity
sR
adjustment

The expected value of future volatility σR is equal to expectation of square root of future variance
è!!!!!!!!!!!
σ@TD = V@TD
è!!!!!!!!!!!
σR = Eσ@TD = E V@TD
¯
Taking second order approximation of square root in V

è!!!!!!!!!!! "###
¯# V@TD − V HV@TD − VL
¯ ¯ 2
V@TD ≈ V + −
¯1ê2 ¯3ê2
2V 8V

And hence
è!!!!!!!!!!! "####
¯ W
σR = E V@TD ≈ V −
¯3ê2
8V