You are on page 1of 45

Timer-Style Options

Design, Pricing and Practice
RiO 2010
Carole Bernard (joint work with Zhenyu Cui)

Carole Bernard

Timer Options

1

Realized Variance

Pricing Timer Options

Comments r = 0%

Theoretical results

Numerical example

Timer-style options

Outline

▶ Realized volatility. What is a timer option?
▶ Model-free price when the risk-free rate is equal to 0.
▶ Some reasons for investing in timer options.
▶ Pricing timer option in general stochastic volatility models Theoretical results.
▶ Pricing timer option in general stochastic volatility models Numerical example.
▶ Further research on timer options.
▶ Other timer-style options and proposal for new timer options.

Carole Bernard

Timer Options

2

Realized Variance

Pricing Timer Options

Comments r = 0%

Theoretical results

Numerical example

Timer-style options

Discrete Realized Variance
∙ Realized Variance over [0, T ] (discretely monitored),
))2
n ( (
1 ∑
Sti
2
Σrealized =
ln
n−1
Sti−1
i=1

where 0 < t1 < ... < tn = T
∙ Annualized Realized Variance over [0, T ] (discretely
monitored),
Σ2
2
= realized . 𝜎
realized
T
∙ Realized variance consumption. As the stock moves daily.
After d days, the “variance budget” is expended according
to
d
2
.
VB realized = 𝜎realized
252
Carole Bernard

Timer Options

3

Realized Variance

Pricing Timer Options

Comments r = 0%

Theoretical results

Numerical example

Timer-style options

Continuous Realized Variance
∙ Assume
)

√ (
dSt
= rdt + Vs 𝜌dWs1 + 1 − 𝜌2 dWs2
St
Realized variance consumption at T (continuously
monitored) or quadratic variation of ln(S) over [0, T ].
∫ T 𝜉
T :=
Vs ds.
0

∙ Let T = nΔ. The cumulative realized variance over [0, T ] is
))2
∫ T
n−1 ( (

S(i+1)Δ
2 𝜉
T = ⟨log (S)⟩T =
(d ln Su ) = lim
ln
n→+∞
SiΔ
0
i=0

∙ (𝜉t )t⩾0 may be viewed as a stochastic clock. We make use
of “time change” techniques.
Carole Bernard

Timer Options

4

Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Modelling the financial market ∙ Assume a constant risk-free interest rate r . Under the risk neutral measure Q (√ ) { √ dSt = rSt dt + Vt St 1 − 𝜌2 dWt1 + 𝜌dWt2 . Carole Bernard Timer Options 5 . and ∫ 𝜉T = T Vt dt 0 is well-defined and converges to +∞ when T → +∞.s. and where 𝛼t and 𝛽t are adapted processes such that a unique solution (St . dVt = 𝛼t dt + 𝛽t dWt2 where W 1 and W 2 are independent Brownian motions. Vt ) exists. Vt > 0 a.

𝜉u = 𝕍} . Vs ds = 𝕍 = inf {u > 0. 0). 0 ∙ The payoff of a timer call option is paid at time 𝜏 and is max(S𝜏 − K.Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Perpetual Timer Options A timer option is a standard option with random maturity. ∙ It is defined as the first hitting time of the realized variance to the variance budget 𝕍 { } ∫ u 𝜏 = inf u > 0. ∙ Denote by 𝜏 the random maturity time of the option. Carole Bernard Timer Options 6 .

Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Pricing timer options ∙ One can write 1 ln(St ) = ln(S0 ) + rt − 𝜉t + 2 ∫ t√ ( ) √ Vs 𝜌dW 2 + 1 − 𝜌2 dW1 0 ∙ Dubins Schwarz theorem applies and one gets ∫ t√ ( ) √ Vs 𝜌dWs2 + 1 − 𝜌2 dWs1 B𝜉t = 0 ∙ One can write 1 ln(St ) = ln(S0 ) + rt − 𝜉t + B𝜉t 2 ∙ Then. Carole Bernard Timer Options 7 . 1 1 S𝜏 = S0 e r 𝜏 e B𝜉𝜏 − 2 𝜉𝜏 = S0 e r 𝜏 e B𝕍 − 2 𝕍 .

) ∫ u √ (√ where B𝕍 = B𝜉(𝜏 ) with Bu = 0 Vt 1 − 𝜌2 dWt1 + 𝜌dWt2 is a Q-standard Brownian motion. 0 . Carole Bernard Timer Options 8 .Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Pricing timer options (cont’d) Theorem The initial price of a timer option is )] [ ( 1 C0 = E Q max S0 e B𝕍 − 2 𝕍 − Ke −r 𝜏 . Not an easy problem because 𝜏 and B𝕍 are not independent and in general B𝕍 ∣𝜏 is not normally distributed.

Thus when r = 0%. K . 0. CBS S0 . T . T𝕍 . 0 . the price of a timer call is equal to ( ) ( ) ˆ1 − K𝒩 d ˆ2 C0∣r=0% = S0 𝒩 d where dˆ1 = Carole Bernard ln ( ) + 21 𝕍 √ 𝕍 S0 K and dˆ2 = dˆ1 − √ 𝕍. Timer Options 9 .Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Pricing timer options when r = 0% When r = 0%. i.e. the price of a timer call option is [ ( )] 1 C0∣r =0% = E Q max S0 e B𝕍 − 2 𝕍 − K . ⇒ closed-form expression equal to the Black and Scholes formula with interest rate r ( = 0%. volatility 𝜎)and maturity T that verify √ 2 𝜎 T = 𝕍.

Carole Bernard Timer Options 10 . 𝜎. ▶ Market Price of the same call CM observed in the market. K . 𝜎. T ) c(𝜎) := CBS (S0 .Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Implied Volatility ▶ Black and Scholes market dSt = rdt + 𝜎dWt St ▶ Call option price CBS (S0 . r . ▶ 𝜎∗ implied volatility such that c(𝜎∗ ) = CM . T ) = S0 𝒩 (d1 ) − Ke −rT 𝒩 (d2 ) where d1 = ln ( S0 K ) ( ) 2 + r + 𝜎2 T √ 𝜎 T √ and d2 = d1 − 𝜎 T . K . r .

ˆ Price at 0 of a standard call option CBS (S0 . Carole Bernard Timer Options 11 . ˆ Price at 0 of a timer call option ( CBS √ S0 . K .T T ) Since the Black and Scholes formula is an increasing function of the volatility. T ) where 𝜎∗ is the implied volatility. one only needs to compare 𝕍 with 𝜎∗2 T . K . 𝕍 .Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options A simple observation Let us compare when r = 0% a standard vanilla call with a perpetual timer call. 𝜎∗ .

Carole Bernard Timer Options 12 . Case 1: If 𝜏 = T . the two options have identical cash-flows.T]) will be larger than the implied volatility at time T for a given option with a strike K then the following strategy is going to make money ▶ Long a standard option ▶ Short a timer option We then investigate what can happen.Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Betting on Realized volatility vs Implied volatility If one believes that the annualized realized volatility (over [0.

then strict profit of K − ST > 0. ▶ If it is negative (S𝜏 > K ) then one sells S and puts K in a bank account. (ST − K )+ − ST + K If ST < K . then 0. the proceeds of the sale will be strictly positive. At time T (since r = 0%). then 𝜏 < T . −(S𝜏 − K )+ ▶ If it is 0 (S𝜏 ⩽ K ) then one makes strict profit by selling the remaining option at time 𝜏 < T . Carole Bernard Timer Options 13 . If ST ⩾ K . P&L ⩾ 0! ▶ At time 𝜏 .Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Cash-flows in Case 2 (𝜏 < T ) If the realized variance 𝜉T > 𝜎∗2 T = 𝕍.

then 𝜏 > T . ▶ If it is positive (ST > K ).Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Cash-flows in Case 3 (𝜏 > T ) If the realized variance 𝜉T < 𝜎∗2 T = 𝕍. (The intuition is that the value of a call option is higher than its exercise value when the underlying does not pay dividends). Carole Bernard Timer Options 14 . Then one would have a strict loss at time T . one receives (ST − K )+ ▶ If it is 0 (ST ⩽ K ) then one has a loss at T because the timer option is still alive with a positive premium. Value at t > (St − K )+ . P&L ⩽ 0! ▶ At time T .

then 𝜏 > T . P&L ⩽ 0! ▶ At time T . Then one would have a strict loss at time T . one receives (ST − K )+ ▶ If it is 0 (ST ⩽ K ) then one has a loss at T because the timer option is still alive with a positive premium. (The intuition is that the value of a call option is higher than its exercise value when the underlying does not pay dividends).Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Cash-flows in Case 3 (𝜏 > T ) If the realized variance 𝜉T < 𝜎∗2 T = 𝕍. Carole Bernard Timer Options 14 . ▶ If it is positive (ST > K ). Value at t > (St − K )+ .

” ⇒ A Long position in a timer option can make money in 80% of the time??? Carole Bernard Timer Options 15 . as well as maturity and strike price. SG CIB calculates that 80% of three-month calls that have matured in-the-money were overpriced. The price of a vanilla call option is determined by the level of implied volatility quoted in the market. reflecting the uncertainty of future market direction. having analyzed all stocks in the Euro Stoxx 50 index since 2000..Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options To conclude. [. this strategy (short position in a timer option) guarantees a sure profit when the realized volatility is higher than the implied volatility! Why is this attractive? Sawyer (2007) explains that “this product is designed to give investors more flexibility and ensure they do not overpay for an option. But the level of implied volatility is often higher than realized volatility.] In fact..

. But the level of implied volatility is often higher than realized volatility. reflecting the uncertainty of future market direction. as well as maturity and strike price.” ⇒ A Long position in a timer option can make money in 80% of the time??? Carole Bernard Timer Options 15 . The price of a vanilla call option is determined by the level of implied volatility quoted in the market. [. SG CIB calculates that 80% of three-month calls that have matured in-the-money were overpriced. this strategy (short position in a timer option) guarantees a sure profit when the realized volatility is higher than the implied volatility! Why is this attractive? Sawyer (2007) explains that “this product is designed to give investors more flexibility and ensure they do not overpay for an option..] In fact. having analyzed all stocks in the Euro Stoxx 50 index since 2000.Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options To conclude.

having analyzed all stocks in the Euro Stoxx 50 index since 2000. this strategy (short position in a timer option) guarantees a sure profit when the realized volatility is higher than the implied volatility! Why is this attractive? Sawyer (2007) explains that “this product is designed to give investors more flexibility and ensure they do not overpay for an option. But the level of implied volatility is often higher than realized volatility. as well as maturity and strike price.] In fact... [.” ⇒ A Long position in a timer option can make money in 80% of the time??? Carole Bernard Timer Options 15 . reflecting the uncertainty of future market direction. The price of a vanilla call option is determined by the level of implied volatility quoted in the market.Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options To conclude. SG CIB calculates that 80% of three-month calls that have matured in-the-money were overpriced.

A. where “perpetual timer options” are called “perpetual mileage options”. 41(4). (1995): “Quadratic-Variation-Based Dynamic Strategies.” Working Paper. ˆ Bick. 722–732. (1990): “Volatility trading.” Management Science. 2008.Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Some references Results and related discussions can be found in ˆ Mileage Options by Neuberger. Carole Bernard Timer Options 16 . London Business School. A. Realized volatility and Mileage options” presented by Roger Lee at the Bachelier meeting. Carr. ˆ “Implied volatility. Joint work with P.

HT and 𝜉T and where f and h are defined by ∫ v √ z 1 dz. √ ST = S0 exp{rT + aT + bT Z}.Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options We assume that the variance process is now modeled by { dVt = 𝛼(Vt ) dt + 𝛽(V(t ) dWt2 ) √ √ dSt = rSt dt + St Vt 1 − 𝜌2 dWt1 + 𝜌dWt2 (1) In the general stochastic volatility model given by (1). h(v ) = 𝛼(v )f ′ (v ) + 𝛽 2 (v )f ′′ (v ) (2) f (v ) = 𝛽(z) 2 0 Carole Bernard Timer Options 17 . bT = (1 − 𝜌2 )𝜉T 2 with ∫ T ∫ T HT = h(Vt )dt and 𝜉T = Vt dt 0 0 and where Z ∼ 𝒩 (0. where aT and bT are defined by 1 aT = 𝜌(f (VT ) − f (V0 )) − 𝜌HT − 𝜉T . 1) independent of VT .

Therefore. Call price = E [Ψ1 (VT . Carole Bernard Timer Options 18 . K . HT ) where 𝜉T := 0 Vs ds and ∫T HT = 0 h(Vs )ds. The price of a standard call option with maturity T is equal to [ ] ˆ0 . 𝜉T .Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Standard European Options Theorem Standard European Call Option. T ) where CBS is( the Black price with ) Scholes call √ b ˆ0 = S0 exp aT + T and 𝜎 S ˆ = bT . HT )] . 𝜎 E CBS (S ˆ. Note that aT and bT 2 ∫T depend on (VT . 𝜉T . r .

0 Timer Options 19 .Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Perpetual Timer Options Theorem Timer Call Option. In a general stochastic volatility model given by (1). ∫ Timer Call Price = E [Ψ2 (𝜏. d2 = d1 − √ (1 − 𝜌2 )𝕍. the price of a timer call option can be calculated as ] [ (1−𝜌2 )𝕍 −r 𝜏 a𝜏 + 2 𝒩 (d1 ) − Ke 𝒩 (d2 ) (3) C0 = E S0 e where a𝜏 = 𝜌(f (V𝜏 ) − f (V0 )) − 𝜌H𝜏 − 12 𝕍 and d1 = ( ) S ln K0 +r 𝜏 +a𝜏 +(1−𝜌2 )𝕍 √ (1−𝜌2 )𝕍 . H𝜏 )] with H𝜏 = Carole Bernard 𝜏 h(Vs )ds. V𝜏 .

𝕍 Timer Call Price = E [Ψ3 (𝜏 )] Carole Bernard Timer Options 20 .Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Perpetual Timer Options (𝜌 = 0) Theorem When 𝜌 = 0. d2 (𝜏 ) = d1 (𝜏 ) − 𝕍. the price of a timer call option in a general stochastic volatility model is given by [ ] C0∣𝜌=0 = S0 E Q [𝒩 (d1 (𝜏 ))] − KE Q e −r 𝜏 𝒩 (d2 (𝜏 )) (4) where ln d1 (𝜏 ) = ( ) S0 K + 12 𝕍 + r 𝜏 √ √ .

H𝜏 ) ∼ 0 𝕍 1 ds. Carole Bernard Timer Options 21 . ∞). V𝜏 . Xs ∫ 0 𝕍 ) h(Xs ) ds Xs where Xt is governed by the SDE { t) df (Xt ) = h(X Xt dt + dBt . then the law of (𝜏. X𝕍 . V𝜏 . H𝜏 ) is given by law (∫ (𝜏. and f and h are given by (2). ∞) 0 is the first passage time of the integrated functional of Vs to the fixed level 𝕍 ∈ (0. Vs ds = 𝕍 ∈ (0.Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Recall that { ∫ t } 𝜏 := 𝜏 (𝕍) = inf t. X0 = V0 (5) (6) where B is a standard Brownian motion.

The estimate of the timer call option price is obtained by Monte Carlo Timer Call Price = E [Ψ2 (𝜏.Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Monte Carlo Approach We will make use of the main result to simulate an i. Timer Call Price when {𝜌 = 0} = E [Ψ3 (𝜏 )] .d. H𝜏 )] . Carole Bernard Timer Options 22 . sample of (𝜏. V𝜏 . H𝜏 ).i. V𝜏 .

We estimate the price by ) ( ˜0 − C0∣r =0% (7) C0mc − 𝜆 C where ˜0 = C S0 e (1−𝜌2 )𝕍 2 n n ∑ i=1 e a𝜏i 𝒩 (d1 (𝜏i . it is possible to significantly improve the convergence of the Monte Carlo estimator.Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Variance Reduction Using the model-free closed-form expression C0∣r =0% for the timer option given in a general stochastic volatility model. H𝜏i ))− n K∑ 𝒩 (d2 (𝜏i . n i=1 ˜0 ) and where a𝜏 . V𝜏i . H𝜏i )) . Carole Bernard Timer Options 23 . V𝜏i . C i (3). d1 and d2 are defined in where 𝜆 = corr (C0mc .

√ dVt = 𝜅 (𝜃 − Vt ) dt + 𝛾 Vt dWt2 where 𝜅. 𝜃 and 𝛾 are the parameters of the volatility process. In the Heston model.s. they are all positive and the Feller condition 2𝜅𝜃 − 𝛾 2 > 0 ensures that Vt > 0 a. Carole Bernard Timer Options 24 . In the Hull and White model.. dVt = aVt dt + 𝜈Vt dWt2 where a and 𝜈 are positive.Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Numerical Example Numerical examples are derived in the Heston model and in the Hull and White model.

04 𝕍 = 0.1 𝜃 = 0.0265 𝜅=2 V0 = 0. Carole Bernard Timer Options 25 .0324 This graph is obtained with a time step of 1/3000 and 1.0625 𝛾 = 0.Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Parameters in the Heston Model Example in the Heston model S0 = 100 K = 100 r = 0. The correlation is 𝜌 = −0. 000 Monte Carlo simulations for each value of the variance budget “𝕍”. 000.5.

1 .04 0.18 16 Timer Call Option when r=4% Timer Call Option when r=0% Timer Call Price 14 12 10 8 6 4 2 0 0.02 0.06 Variance Budget 0.08 0.

ˆ Second Figure: implementing our techniques using 50. Vt ) to get the price of a timer option by Monte Carlo as E [e −r 𝜏 (S𝜏 − K )+ ] The number in parenthesis are the standard deviations obtained with 100. ˆ First Figure: direct simulation of (St . Pay attention to the scales of the y-axis. Carole Bernard Timer Options 27 .000 simulations.Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options The transformation we do is good because it allows us to get a much better convergence.000 simulations for each value of M.

8 Price when ρ=0 8 7.5 9 ρ = − 0.5 7 6.5 6 0 ρ=0 ρ = 0.5 ρ=−0.8 50 100 150 Number of discretization step 200 .8 ρ=0 ρ=0.9.8 Price 8.

45 7.8 7.6 7.5 7.8 ρ=−0.7 Price 7.55 7.85 7.75 7.7.8 100 200 300 400 500 600 700 800 Number of discretization step 900 1000 .4 7.35 0 ρ=0.65 ρ=0 7.

Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Further issues Specific questions related to the timer options ˆ Which scheme is efficient to similate the Bessel process underlying the price of a timer option? ˆ Can we use results on Bessel processes to improve the pricing and hedging of timer options? More general questions ˆ What about if the realized variance is discretely monitored instead of continuously monitored? ˆ What about the effect of stochastic interest rates and correlation between interest rates and volatility? Carole Bernard Timer Options 30 .

Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Further issues Specific questions related to the timer options ˆ Which scheme is efficient to similate the Bessel process underlying the price of a timer option? ˆ Can we use results on Bessel processes to improve the pricing and hedging of timer options? More general questions ˆ What about if the realized variance is discretely monitored instead of continuously monitored? ˆ What about the effect of stochastic interest rates and correlation between interest rates and volatility? Carole Bernard Timer Options 30 .

Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Further issues Specific questions related to the timer options ˆ Which scheme is efficient to similate the Bessel process underlying the price of a timer option? ˆ Can we use results on Bessel processes to improve the pricing and hedging of timer options? More general questions ˆ What about if the realized variance is discretely monitored instead of continuously monitored? ˆ What about the effect of stochastic interest rates and correlation between interest rates and volatility? Carole Bernard Timer Options 30 .

07 0.8 Timer Call Option when ρ=0 Timer Call Option when ρ=0.02 0.9 Timer Call Price 8.04 0.08 .03 0.06 0.8 6.5 8 7.5 7 Timer Call Option when ρ=−0.01 0.05 Interest rate r 0.5 6 0 0.

Carole Bernard Timer Options 32 . T )? How large is the difference relative to the perpetual timer options? EQ [e −r 𝜏 1𝜏 <T (S𝜏 − K )+ + e −rT 1𝜏 ≥T (ST − K )+ ].Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Finite Expiry Timer Options How to value a timer option with expiry min(𝜏.

Carole Bernard Timer Options 33 . Realized volatility and Mileage options” presented by Roger Lee at the Bachelier meeting. 2008. Joint work with P.Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Finite Expiry Timer Options A model-free upper bound can be derived [ ] EQ e −r 𝜏 ∧T (S𝜏 ∧T − K )+ [ ] [ ] ⩽ min(EQ e −r 𝜏 (S𝜏 − K )+ . Carr. EQ e −rT (ST − K )+ ) “Implied volatility.

Carole Bernard Timer Options 34 . In a FX timer option. The only difference lies in the underlying process.Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options FX Timer options A FX timer option is very similar to a timer call. When the accumulated variance of the exchange rate has reached the budget. the caplet expires. the investor sets a variance budget for the exchange rate and it is a call option on the exchange rate (“timer caplet”).

on the other hand.Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Time Swap (description) Sawyer (2007). In other words. Rather than volatility. the strike is expressed in days. gives investors a means to short volatility with an inverted convexity profile (meaning the downside is limited). if an investor wants to short volatility over a specified investment period . three months . it will take longer than three months for the option to expire. if realised volatility is consistently lower than the specified level. and the investor receives a payout.a variance budget is calculated with a volatility level set by SG CIB. times the notional.” Carole Bernard Timer Options 35 . “The time swap.for instance. So. The payout is based on the number of days required to consume the variance budget minus the specified investment horizon.

the payoff is K (𝜏 − T ) . the price of the time swap is given as follows [ ] KE e −r 𝜏 (𝜏 − T ) . ▶ At 𝜏 when the variance budget is expended. a time swap has a notional amount. K . 0 ▶ As a standard swap. Then. Vs ds = 𝕍}.Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Time Swap (Modelling) ▶ Compare the target expiry time T with the random time 𝜏 defined as ∫ u 𝜏 = inf{u > 0. ▶ Some properties of a long position ˆ positive payoff if 𝜏 > T and negative when 𝜏 < T . ˆ short position in realized volatility. ˆ limited exposure (maximum loss is KT ) Carole Bernard Timer Options 36 .

Sawyer explains that “the out-performance product is similar to the timer call the investor specifies a target volatility for the spread between two underlyings and a target investment horizon. Mattatia claims the timer out-performance call can be 30 percent cheaper than a plain vanilla out-performance option ” Carole Bernard Timer Options 37 .Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Timer out-performance option The timer out-performance option was developed shortly after the first timer call option was sold in April 2007. which is used to calculate a variance budget.

u ˆ Payoff at 𝜏 of the timer out-performance option max (S2 (𝜏 ) − S1 (𝜏 ). 0) . ˆ A variance budget is calculated as 𝕍 = 𝜎02 T .Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Timer out-performance option (modelling) ˆ Consider two correlated assets S1 and S2 . ln S1 } 〉 =𝕍 . Define then { 𝜏 〈 S2 = inf u > 0. Carole Bernard Timer Options 38 . ˆ Specify a target volatility 𝜎0 for the spread between the two underlying’s log-return and a target investment horizon T .

.Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options Timer-style options ˆ Forward start timer option ˆ Compound timer option ˆ Timer cliquet option ˆ Timer barrier option ˆ Timer Lookback option ˆ .. Carole Bernard Timer Options 39 . Discretization is not uniformly done but done along a random grid.

Columbia University. London Business School. 179–207. N.. ˆ Lee. 20(7). 319–339. ˆ Carr.Risk Awards 2008.Realized Variance Pricing Timer Options Comments r = 0% Theoretical results Numerical example Timer-style options ˆ Bernard C. (1990): “Volatility trading. R. 2008. 1.” RISK magazine. 21(1). Realized volatility and Mileage options”. and Lee. (2009): “Volatility Derivatives. P. Bachelier meeting. forthcoming.” PhD thesis. (2008): “Equity Derivatives House of the Year . P. ˆ Neuberger. ˆ Carr. Carole Bernard Timer Options 40 . R. (2010): “Managing Volatility Risk. 14. Carr. ˆ Sawyer. Joint work with P. (2008): “Implied volatility.X.Soci´et´e G´en´erale . ˆ Sawyer.” RISK magazine.” Working Paper.” Power Point.” Finance & Stochastics. R. and Cui Z.. ˆ Soci´et´e G´en´erale Corporate & Investment Banking (2007): “The Timer Call: A Floating Maturity Option. Journal of Computational Finance. C. (2010): “Hedging Variance Options on Continuous Semimartingales. A.” Annual Review of Financial Economics. and Lee. (2010): “Pricing Timer Options”. ˆ Li. (2007): “SG CIB Launches Timer Options. N.