FINA 276 OPTIONS - PAPER: GREEKS DERIVATION FROM

BLACK-SCHOLES FORMULA

ARGYN KUKETAYEV

Abstract. We derive formulae for Greeks, then use them to analyse European call and
put option price sensitivity to input parameters. Only non-dividend paying stocks are
considered.

1. Black-Scholes Option Price Formula

Option price formulae (non-dividend paying stocks) are given in equation 12.20 of [1] for
call option:
c = S0 N (d1 ) − Ke−rT N (d2 ) (1.1)
and in equation 12.21 for put option
p = Ke−rT N (−d2 ) − S0 N (−d1 ) (1.2)
where
ln(S0 /K) + (r + σ 2 /2)T
d1 = √ (1.3)
σ T
ln(S0 /K) + (r − σ 2 /2)T √
d2 = √ = d1 − σ T (1.4)
σ T
Here, N (x) is a cumulative normal distribution function, which is defined as
Z x
N (x) = Z(t)dt (1.5)
−∞

where
1 2
Z(x) = √ e−x /2 (1.6)
2π
see [2] equations 26.2.1 and 26.2.2.
In next five sections we shall get the formulae for five Greeks: delta, theta, gamma, vega
and rho.

Date: March 20, 2007.

1991 Mathematics Subject Classification. Primary 91B28 , Secondary 91B70 .
Key words and phrases. black scholes greek gamma delta theta sensitivity vega rho.
1
2 ARGYN KUKETAYEV

2. Delta
2.1. Call Option Delta.
Definition 1. Delta is defined by the following equation (see 14.4 in [1])
∂c
∆= (2.1)
∂S
Substituting the price with 1.1
∂ 
SN (d1 ) − Ke−rT N (d2 )

∆= (2.2)
∂S
∂N (d1 ) ∂N (d2 )
= N (d1 ) + S − Ke−rT (2.3)
∂S ∂S
∂N (d2 )
Let’s evaluate the last term in the above equation using 2.14 for ∂S

∂N (d2 ) Z(d1 )(S/K)erT Z(d1 )

Ke−rT = Ke−rT √ = √ (2.4)
∂S σ TS σ T
∂N (d1 )
then put the obtained formula back in 2.3 using also 2.13 for ∂S

Z(d1 ) Z(d1 )
∆ = N (d1 ) + S √ − √
σ TS σ T
Z(d1 ) Z(d1 )
= N (d1 ) + √ − √
σ T σ T
Finally,
∆ = N (d1 ) (2.5)

2.2. Put Option Delta.

Definition 2. Delta is defined by the following equation (see 14.4 in [1])
∂p
∆= (2.6)
∂S
Substituting the price with 1.2
∂  −rT
∆= Ke N (−d2 ) − SN (−d1 ) (2.7)
∂S
∂N (−d1 ) ∂N (−d2 )
= −N (−d1 ) − S + Ke−rT (2.8)
∂S ∂S
Let’s evaluate the last term in the above equation using 2.16 for ∂N∂S
(−d2 )

S rT
∂N (−d2 ) Z(−d1 ) K e Z(−d1 )
Ke−rT = −Ke−rT √ =− √ (2.9)
∂S σ TS σ T
 FINA 276: GREEKS DERIVATION 3

∂N (−d1 )
then put the obtained formula back in 2.8 using also 2.15 for ∂S
Z(−d1 ) Z(−d1 )
∆ = −N (−d1 ) + S √ − √ = −N (−d1 )
σ TS σ T
Finally, using the equation N (x) + N (−x) = 1, which can be derived from 26.2.5 and 26.2.6
in [2], we get
∆ = N (d1 ) − 1 (2.10)
∂N (d1 ) ∂N (d2 )
2.3. Auxilliary expressions. Let’s obtain forumale for ∂S
and ∂S
.

∂d1 ∂ ln(S/K) + (r + σ 2 /2)T

= √ (2.11)
∂S ∂S σ T
∂d1 1 ∂ 1 1 1 1
= √ ln(S/K) = √ = √ (2.12)
∂S σ T ∂S σ T (S/K) K σ TS
Therefore,
∂N (d1 ) ∂d1 ∂d1 Z(d1 )
= N 0 (d1) = Z(d1 ) = √ (2.13)
∂S ∂S ∂S σ TS
and also,
√
∂N (d2 ) 0 ∂d2 0 ∂[d1 − σ T ] ∂d1 Z(d2 )
= N (d2 ) = N (d2 ) = N 0 (d2 ) = √
∂S ∂S ∂S ∂S σ TS
Finally, using 8.3
∂N (d2 ) Z(d1 )(S/K)erT
= √ (2.14)
∂S σ TS
∂N (−d1 ) ∂N (−d2 )
Now, let’s obtain forumale for ∂S
and ∂S
. Using 2.12 we can derive the following
∂N (−d1 ) ∂d1 ∂d1 Z(−d1 )
= −N 0 (−d1) = −Z(−d1 ) =− √ (2.15)
∂S ∂S ∂S σ TS
Similarly,
√
∂N (−d2 ) ∂d 2 ∂[d1 − σ T] ∂d1 Z(−d2 )
= −N 0 (−d2 ) = −N 0 (−d2 ) = −N 0 (−d2 ) =− √
∂S ∂S ∂S ∂S σ TS
then, using 8.4
∂N (−d2 ) Z(−d1 )(S/K)erT
=− √ (2.16)
∂S σ TS
3. Gamma
Both call and put options have the same expression for Γ. We’ll prove this statement in
next two subsections.
Z(d1 )
Γ= √ (3.1)
σ TS
4 ARGYN KUKETAYEV

3.1. Call Option Gamma.

Definition 3. Gamma is defined by equation (see 14.6 in [1])
 
∂ ∂c
Γ= (3.2)
∂S ∂S
Hence, using definition of ∆ (2.1) and its value (2.5),
∂∆ ∂N (d1 )
Γ= = (3.3)
∂S ∂S
∂N (d1 )
Finally, use 2.13 to substitute ∂S
∂N (d1 ) Z(d1 )
Γ= = √ (3.4)
∂S σ TS
3.2. Put Option Gamma.
Definition 4. Gamma is defined by equation (see 14.6 in [1])
 
∂ ∂p
Γ= (3.5)
∂S ∂S
Hence, using definition of ∆ (2.6) and its value (2.10),
∂∆ ∂(N (d1 ) − 1) ∂N (d1 )
Γ= = = (3.6)
∂S ∂S ∂S
You can see that put option has the same Γ as call options (look at equation 3.3).

4. Theta
In 14.5 of [1], Θ is defined as a ”the rate of change of the value of the portfolio with respect
to the passage of time with all else remaining the same”.
4.1. Call Option Theta.
Definition 5. We re-define Θ as follows
∂c
Θ=− (4.1)
∂T
Note the sign. In Black-Scholes formula 1.1 the option price is expressed as a function
of time to maturity T . Since we want to measure sensitivity with respect to passing (or
current) time t, then the sign must be negative, because δt = −δT .
Let’s substitute c in 4.1 by Black-Scholes formula 1.1
∂ 
SN (d1 ) − Ke−rT N (d2 )

Θ=− (4.2)
∂T
  
∂N (d1 ) ∂ −rT −rT ∂
=− S − K N (d2 ) e +e N (d2 ) (4.3)
∂T ∂T ∂T
 FINA 276: GREEKS DERIVATION 5
  
∂N (d1 ) −rT −rT ∂
=− S − K −re N (d2 ) + e N (d2 ) (4.4)
∂T ∂T
  
∂N (d1 ) −rT ∂
=− S − Ke −rN (d2 ) + N (d2 ) (4.5)
∂T ∂T
Use 4.29 and 4.28 for ∂N∂T(d1 ) and ∂N∂T(d2 )
  
d2 r −rT rT d1 r
= − −Z(d1 )( − √ )S − Ke −rN (d2 ) − Z(d1 )(S/K)e ( − √ ) (4.6)
2T σ T 2T σ T
 
d2 r d1 r −rT
= − −Z(d1 )( − √ )S + Z(d1 )( − √ )S + Ke rN (d2 ) (4.7)
2T σ T 2T σ T
 
d1 − d2 −rT
= − Z(d1 ) S + Ke rN (d2 ) (4.8)
2T
Finally,  
σ −rT
Θ = − Z(d1 )S √ + Kre N (d2 ) (4.9)
2 T
4.2. Put Option Theta. Just like for call options in see 4.1 let’s define Θ as follows
Definition 6.
∂p
Θ=− (4.10)
∂T
Let’s substitute p in 4.10 by Black-Scholes formula 1.2
∂ 
−SN (−d1 ) + Ke−rT N (−d2 )

Θ=− (4.11)
∂T
  
∂N (−d1 ) ∂ −rT −rT ∂
= − −S + K N (−d2 ) e +e N (−d2 ) (4.12)
∂T ∂T ∂T
  
∂N (−d1 ) −rT −rT ∂
= − −S + K −re N (−d2 ) + e N (−d2 ) (4.13)
∂T ∂T
  
∂N (−d1 ) −rT ∂
= − −S + Ke −rN (−d2 ) + N (−d2 ) (4.14)
∂T ∂T
Use 4.31 and 4.30 for ∂N∂T
(−d1 )
and ∂N∂T
(−d2 )

  
d2 r −rT S rT d1 r
= − −Z(d1 )( − √ )S + Ke −rN (−d2 ) + Z(d1 ) e ( − √ ) (4.15)
2T σ T K 2T σ T
 
d2 r d1 r −rT
= − −Z(d1 )( − √ )S + Z(d1 )( − √ )S − Ke rN (−d2 ) (4.16)
2T σ T 2T σ T
 
d1 − d2 −rT
= − Z(d1 ) S − Ke rN (−d2 ) (4.17)
2T
6 ARGYN KUKETAYEV

Finally,  
σ −rT
Θ = − Z(d1 )S √ − Kre N (−d2 ) (4.18)
2 T
∂N (d1 ) ∂N (d2 )
4.3. Auxilliary expressions. Let’s obtain formulae for ∂T
and ∂T
. We shall need
expressions for ∂d
∂T
1
and ∂d
∂T
2

∂d1 ∂ ln(S/K) + (r + σ 2 /2)T

= √ (4.19)
∂T ∂T σ T
σ2 σ2
   
1 ∂ S S ∂ 1
= √ ln( ) + (r + )T + ln( ) + (r + )T √ (4.20)
σ T ∂T K 2 K 2 ∂T σ T
σ2 σ2
  
1 S 1 1
= √ (r + ) + ln( ) + (r + )T − √ (4.21)
σ T 2 K 2 2 σT T
S S 2
1 1 σ2 1 ln( K ) 1 ln( K ) − (r + σ2 )T
= √ (r + ) − √ =− √ (4.22)
2σ T 2 2 σT T 2T σ T
S 2
1 ln( K ) + (r − σ2 )T − 2rT d2 r
=− √ =− + √ (4.23)
2T σ T 2T σ T
∂d1 d2 r
=− + √ (4.24)
∂T 2T σ T
and √
∂d2 ∂ √ ∂d1 ∂ T ∂d1 1 σ
= (d1 − σ T ) = −σ = − √ (4.25)
∂T ∂T ∂T ∂T ∂T 2 T
√
d2 r 1 σ d1 − σ T 1 σ r d1 r
=− + √ − √ =− − √ + √ =− + √ (4.26)
2T σ T 2 T 2T 2 T σ T 2T σ T
∂d2 d1 r
=− + √ (4.27)
∂T 2T σ T
1
Finally, using 4.24 and 4.27
   
∂N (d2 ) 0 ∂d2 d1 r S rT d1 r
= N (d2 ) = −Z(d2 ) − √ = −Z(d1 ) e − √ (4.28)
∂T ∂T 2T σ T K 2T σ T
 
∂N (d1 ) ∂d1 d2 r
= N 0 (d1 ) = −Z(d1 ) − √ (4.29)
∂T ∂T 2T σ T
∂N (−d1 ) ∂N (−d2 )
Let’s obtain formulae for ∂T and ∂T .
   
∂N (−d2 ) 0 ∂d2 d1 r S rT d1 r
= −N (−d2 ) = Z(−d2 ) − √ = Z(d1 ) e − √ (4.30)
∂T ∂T 2T σ T K 2T σ T
1Note,that we think that in [3] a partial derivative ∂d d2
∂T is evaluated incorrectly to − 2T . When last time
1

we checked, i.e. on March 20, 2007, this error was not corrected yet. We believe that they mistakingly
substituted a fraction in 4.22 by d2. Subsequently, a partial derivative ∂d ∂T is also evaluated incorrectly.
2

However, these errors ”cancel” each other, and the final expression for Θ is correct.
 FINA 276: GREEKS DERIVATION 7
 
∂N (−d1 ) ∂d1 d2 r
= −N 0 (−d1 ) = Z(d1 ) − √ (4.31)
∂T ∂T 2T σ T
5. Vega
Both call and put options have the same Vega. First, we’ll show the equation, then we’ll
prove it in next two subsections.2
√
Υ = Z(d1 )S T (5.1)
5.1. Call Option Vega.
Definition 7. Vega is defined as follows (see 14.8 in [1]).
∂c
Υ= (5.2)
∂σ
Substitute c with its formula 1.1
∂ 
SN (d1 ) − Ke−rT N (d2 )

Υ= (5.3)
∂σ
∂N (d1 ) ∂
=S − Ke−rT N (d2 ) (5.4)
∂σ ∂σ
Now, using 5.21 and 5.20 for ∂N∂σ
(d1 )
and ∂N∂σ
(d2 )

 
d2 −rT S rT d1
= −SZ(d1 ) − Ke (−Z(d1 )) e (5.5)
σ K σ
d2 d1 d1 − d2 √
= −SZ(d1 ) + Z(d1 )S = SZ(d1 ) = Z(d1 )S T (5.6)
σ σ σ
5.2. Put Option Vega.
Definition 8. Vega is edfined as follows (see 14.8 in [1]).
∂c
Υ= (5.7)
∂σ
Substitute p with its formula 1.2
∂ 
−SN (−d1 ) + Ke−rT N (−d2 )

Υ= (5.8)
∂σ
∂N (−d1 ) ∂
= −S + Ke−rT N (−d2 ) (5.9)
∂σ ∂σ
Now, using 5.23 and 5.22 for ∂N ∂σ
(−d1 )
and ∂N ∂σ
(−d2 )

 
d2 −rT S rT d1
= −SZ(d1 ) + Ke Z(d1 ) e (5.10)
σ K σ
2I didn’t know which greek letter to use for Vega and chose upsilon.
8 ARGYN KUKETAYEV

d2 d1 d1 − d2 √
= −SZ(d1 ) + Z(d1 )S = SZ(d1 ) = Z(d1 )S T (5.11)
σ σ σ
∂N (d1 ) ∂N (d2 )
5.3. Auxilliary expressions. Let’s obtain forumale for ∂σ
and ∂σ
.
First, let’get ∂d
∂σ
1
and ∂d
∂σ
2

∂d1 ∂ ln(S/K) + (r + σ 2 /2)T

= √ (5.12)
∂σ ∂σ σ T
σ2 σ2
   
1 ∂ S S ∂ 1
= √ ln( ) + (r + )T + ln( ) + (r + )T √ (5.13)
σ T ∂σ K 2 K 2 ∂σ σ T
σ2
  
1 S 1 1
= √ σT + ln( ) + (r + )T − 2 √ (5.14)
σ T K 2 σ T
σ2
     
1 2 S 1 1
= − − √ σ T + ln( ) + (r + )T − 2 √ (5.15)
2
σ T K 2 σ T
σ2
   
S 1 1
= ln( ) + (r − )T − 2 √ (5.16)
K 2 σ T
S σ2
1 ln( K ) + (r − 2
)T
=− √ (5.17)
σ σ T
Therefore,
∂d1 d2
=− (5.18)
∂σ σ
and
√
∂d2 ∂ √ ∂d1 √ d2 √ d2 + σ T d1
= (d1 − σ T ) = − T =− − T =− =− (5.19)
∂σ ∂σ ∂σ σ σ σ
Finally, using 5.18 and 5.19
∂ ∂d2 d1 d1
N (d2 ) = N 0 (d2 ) = −Z(d2 ) = −Z(d1 )(S/K)erT (5.20)
∂σ ∂σ σ σ
∂ ∂d1 d2
N (d1 ) = N 0 (d1 ) = −Z(d1 ) (5.21)
∂σ ∂σ σ
∂N (−d1 ) ∂N (−d2 )
Let’s obtain forumale for ∂σ
and ∂σ
using 5.18 and 5.19.
∂ ∂d2 d1 d1
N (−d2 ) = −N 0 (−d2 ) = Z(d2 ) = Z(d1 )(S/K)erT (5.22)
∂σ ∂σ σ σ
∂ ∂d1 d2
N (−d1 ) = −N 0 (−d1 ) = Z(d1 ) (5.23)
∂σ ∂σ σ
 FINA 276: GREEKS DERIVATION 9

6. Rho
6.1. Call Option Rho.
Definition 9. Rho is defined by equation (see 14.9 in [1])
∂c
rho = (6.1)
∂r
Substitute c with its formula 1.1
∂ 
SN (d1 ) − Ke−rT N (d2 )

rho = (6.2)
∂r
 
∂N (d1 ) ∂ −rT −rT ∂
=S − K N (d2 ) e +e N (d2 )
∂r ∂r ∂r
 
∂N (d1 ) −rT ∂
=S − Ke −T N (d2 ) + N (d2 ) (6.3)
∂r ∂r
∂N (d1 ) ∂N (d2 )
Using 6.15 and 6.16 for ∂r
and ∂r
√ √ !
T S T
= SZ(d1 ) − Ke−rT −T N (d2 ) + Z(d1 ) erT (6.4)
σ K σ
√ √
T −rT T
= SZ(d1 ) + Ke T N (d2 ) − SZ(d1 ) (6.5)
σ σ
Finally,
rho = Ke−rT T N (d2 ) (6.6)
6.2. Put Option Rho.
Definition 10. Rho is defined by equation (see 14.9 in [1])
∂p
rho = (6.7)
∂r
Substitute p with its formula 1.2
∂ 
−SN (−d1 ) + Ke−rT N (−d2 )

rho = (6.8)
∂r
 
∂N (−d1 ) ∂ −rT −rT ∂
= −S + K N (−d2 ) e +e N (−d2 )
∂r ∂r ∂r
 
∂N (−d1 ) −rT ∂
= −S + Ke −T N (−d2 ) + N (−d2 ) (6.9)
∂r ∂r
∂N (−d1 ) ∂N (−d2 )
Using 6.17 and 6.18 for ∂r
and ∂r
√ √ !
T S rT T
= SZ(d1 ) + Ke−rT −T N (−d2 ) − Z(d1 ) e (6.10)
σ K σ
10 ARGYN KUKETAYEV
√ √
T −rT T
= SZ(d1 ) − Ke T N (−d2 ) − SZ(d1 ) (6.11)
σ σ
Finally,
rho = −Ke−rT T N (−d2 ) (6.12)
6.3. Auxilliary expressions. Let’s obtain forumale for ∂N∂r(d1 ) and ∂N∂r(d2 ) . First, we need
∂d1
and ∂d 2
∂r ∂r √
∂d1 ∂ ln(S/K) + (r + σ 2 /2)T T
= √ = (6.13)
∂r ∂r σ T σ
√
∂d2 ∂ √ ∂d1 T
= (d1 − σ T ) = = (6.14)
∂r ∂σ ∂σ σ
Finally, √
∂ 0 ∂d1 T
N (d1 ) = N (d1 ) = Z(d1 ) (6.15)
∂r ∂r σ
then using 8.3 √ √
∂ ∂d 2 T S T
N (d2 ) = N 0 (d2 ) = Z(d2 ) = Z(d1 ) erT (6.16)
∂r ∂r σ K σ
Let’s obtain forumale for ∂N (−d
∂r
1)
and ∂N (−d
∂r
2)
using equations for ∂d
∂r
1
and ∂d
∂r
2
.
√
∂ ∂d1 T
N (−d1 ) = −N 0 (−d1 ) = −Z(d1 ) (6.17)
∂r ∂r σ
then using 8.3
√ √
∂ 0 ∂d2 T S rT T
N (−d2 ) = −N (−d2 ) = −Z(d2 ) = −Z(d1 ) e (6.18)
∂r ∂r σ K σ
7. Summary and sensitivity analysis
7.1. Call Option Sensitivity. Let’s summarize the obtained results (2.5, 3.1, 4.9, 5.1 and
6.6) in a Table 1.
Table 1. Greeks for non-dividend paying call options

Greek Definition Formula Sign

∂c
Delta (∆) ∂S

N (d1√
) ≥0
∂ ∂c
Gamma (Γ) ∂S ∂S n Z(d1 )/(σ T S) o ≥0
∂c σ −rT
Theta (Θ) − ∂T − Z(d1 )S 2√T + Kre N (d2 ) ≤0
∂c
√
Vega ∂σ
Z(d1 )S T ≥0
Rho ∂c
∂r
Ke−rT T N (d2 ) ≥0

The last column shows the sign of the Greek. For example, Θ has a non-positive value
at any combination of its parameters. It’s easy to see where these signs come from if you
remember that N (x), Z(x) and ex are positive at any x.
 FINA 276: GREEKS DERIVATION 11

Now we have almost all inputs to prove the call option part of the sensitivity Table 2.
These results can be explained by using Taylor series expansion (see 9.12 in [2]). Consider
the first order term of the Taylor series for asset price changes:
∂c
cS0 +δS = c0 + · δS + ... = c0 + ∆ · δS + ...
∂S

Table 2. Call option price sensitivity to inputs

Input Impact Explanation

Asset price, S + ∆≥0
Time (current), t = T0 − T - Θ≤0
Time to maturity, T + −Θ ≥ 0
Volatility, σ + V ega ≥ 0
Risk free rate, r + Rho ≥ 0

In order to analyze a sensitivity to changing time, let’s suppose that the current time is
t0 , option value at time t0 + δt is
∂c
ct0 +δt = ct0 + · δt + ...
∂t
∂c ∂c
Knowing that δt = −δT we have ∂t
= − ∂T . Therefore,
∂c
ct0 +δt = ct0 − · δt + ... = ct0 + Θ · δt + ...
∂T
i.e. as time passes by European call option value tends to decrease because Θ is a negative
number. On the other hand, if time to maturity increases, then option value should increase
too.
Sensitivity to volatility and risk free rate changes can be analyzed similarly. Increased
volatility and risk free rate should increase European call option price.

7.2. Put Option Sensitivity. Just like for call options, let’s summarize our findings (2.10,
3.1, 4.18, 5.1 and 6.12) in a Table 3.

Table 3. Greeks for non-dividend paying put options

Greek Definition Formula Sign

∂p
Delta (∆) ∂S

−1
N (d1 ) √ ≤0
∂ ∂p
Gamma (Γ) ∂S ∂S n Z(d1 )/(σ T S) o ≥0
∂c σ −rT
Theta (Θ) − ∂T − Z(d1 )S 2√T − Kre N (−d2 ) ???
∂p
√
Vega ∂σ
Z(d1 )S T ≥0
∂p −rT
Rho ∂r
−Ke T N (−d2 ) ≤0
12 ARGYN KUKETAYEV

Our analysis is similar to call option sensitivity analysis. In fact, the second and fourth
rows are exactly the same. Let’s look at the differences.
First of all, ∆ is non-positive for put options. It’s easy to see why if you recall that
N (x) is the cumulative normal distribution function, i.e. it can’t be greater than 1. Hence,
N (x) − 1 ≤ 0 for any given x.
Second, Rho is non-positive for put options. Look at the expression for Rho, all its factors
(K, ex , T and N (x)) are positive, therefore Rho can not be positive.
The issue is with Θ, because its expression has a subtraction inside Z(d1 )S 2√σT −Kre−rT N (−d2 ).
Depending on the inputs, it can be positive or negative.

Table 4. Put option price sensitivity to inputs

Input Impact Explanation

Asset price, S - ∆ ≤ 0S
Time (current), t = T0 − T ??? Θ ≥ 0SΘ ≤ 0
Time to maturity, T ??? Θ≥0 Θ≤0
Volatility, σ + V ega ≥ 0
Risk free rate, r - Rho ≤ 0

Now we have almost all inputs to prove the put option part of the sensitivity Table 4. The
results can be explained by using Taylor series expansion (see 9.12 in [2]). Consider the first
order of the Taylor series for asset price changes:
∂c
cS0 +δS = c0 + · δS + ... = c0 + ∆ · δS + ...
∂S
Put option has non-positive ∆. Therefore, increase in stock price should decrease the
option price. Sensitivity to volatility and risk free rate changes can be analyzed similarly.
Increased volatility and decreased risk free rate should increase European put option price.
Due to the fact that Θ can be positive or negative, put option price’s sensitivity to time
to maturity is uncertain in general case, its sign depends on inputs.

8. Common expressions
By definition (see 26.1 in [2])
N 0 (x) = Z(x) (8.1)
Hence,
x 2
N 00 (x) = Z 0 (x) = − √ e−x /2 (8.2)
2π
Let’s show the relation between Z(d2 ) and Z(d1 ).
1 √ 2 1 2 2
√
Z(d2 ) = √ e−(d1 −σ T ) /2 = √ e−d1 /2−(σ T −2d1 σ T )/2
2π 2π
 FINA 276: GREEKS DERIVATION 13
2
Remember that Z(d1 ) = √1 e−d1 /2 and substitute it in the above formula
2π
2 T −2d σ
√
= Z(d1 )e−(σ 1 T )/2

then expand d1
ln(S/K)+(r+σ 2 /2)T √
−σ 2 T /2+ √ σ T 2 2 /2)T
= Z(d1 )e σ T = Z(d1 )e−σ T /2+ln(S/K)+(r+σ
σ2 T σ2 T
n o
= Z(d1 ) eln(S/K) ert e− 2 + 2 = Z(d1 )(S/K)erT
Finally,
Z(d2 ) = Z(d1 )(S/K)erT (8.3)
Now, using this last equation we can get the following relation between Z(−d2 ) and
Z(−d1 ).
Z(−d2 ) = Z(d2 ) = Z(d1 )(S/K)erT = Z(−d1 )(S/K)erT (8.4)
Conclusion 1. We derived five Greeks from Black-Scholes equation for European call and
put options with non-divident paying stocks. We analyzed the sensitivity of option prices
to changes in input parameters, and were able to figure out the sign of the impact of such
changes in asset prices, volatility and risk free rate. We identified that decreasing time to
maturity should increase call option price. However, for put options it could either decrease
or increase the option price depending on the input parameters.
We didn’t look at the sensitivity to strike price and dividends changes, because it was out
of the scope of this work. The focus of this excersize was to derive Greeks from Black-Scholes
formula.
References
[1] John C. Hull Options, Futures and Other Derivatives, 5th Ed., (Prentice Hall)
[2] Edited by Milton Abramowitz and Irene A. Stegun, Handbook of Mathematical Functions, 10th Ed.
[3] Henry Tang, The Greek Letters - Theta, Quantnotes.com, http://quantnotes.com/fundamentals/options/thegreeks-
theta.htm
E-mail address, Argyn Kuketayev: jawabean@gwu.edu