Michael Brenndoerfer | Data & AI, Private Equity, Technology

Michael Brenndoerfer

The Greeks and Option Risk Management › Greeks as a Taylor Expansion

Michael BrenndoerferNovember 29, 20256 min read

Quantitative FinanceAll chapters

Part 5 of 15Stochastic Calculus and Pricing Theory

Chapter 31The Greeks and Option Risk Management › Greeks as a Taylor Expansion

Reading section 7 of 14 · Greeks as a Taylor Expansion · 6 minSection 7 of 14 · 6 min

Section 7 of 14 · 6 min read

The Greeks and Option Risk Management › Greeks as a Taylor ExpansionOne section · Sections in this chapter7 / 14 ↓

Sections in this chapter14

Reading tools Definitions & inline links

Choose your expertise level to adjust how many terms are explained. Beginners see more tooltips, experts see fewer to maintain reading flow. Hover over underlined terms for instant definitions.

Article links

Make inline references clickable

The Greeks provide a powerful way to approximate how an option's value will change when multiple market parameters shift simultaneously. This approximation is based on a Taylor series expansion, the same mathematical tool we developed in Part I for approximating functions near a point. The Taylor expansion allows us to decompose complex changes in option value into contributions from each risk factor.

First-Order ApproximationLink Copied

For small changes in the underlying inputs, we can write:

dV \approx \Delta \cdot dS + \Theta \cdot dt + \nu \cdot d\sigma + \rho \cdot dr

where:

$dV$ : change in option price
$\Delta$ : option delta
$dS$ : change in stock price
$dt$ : passage of time
$d\sigma$ : change in implied volatility
$dr$ : change in interest rate
$\Theta$ : option theta
$\nu$ : option vega
$\rho$ : option rho

This first-order approximation is linear in each parameter and forms the basis of daily P&L attribution in trading operations. Each Greek acts as a coefficient that translates a change in its corresponding market variable into a contribution to profit or loss. By examining these contributions separately, we can understand why our positions made or lost money and can identify which risk factors drove the result.

Second-Order ApproximationLink Copied

To improve accuracy, especially for larger moves, we include the second-order term:

dV \approx \Delta \cdot dS + \frac{1}{2}\Gamma \cdot (dS)^2 + \Theta \cdot dt + \nu \cdot d\sigma + \rho \cdot dr

where:

$dV$ : change in option price
$\Delta$ : option delta
$dS$ : change in stock price
$\Gamma$ : option gamma
$(dS)^2$ : squared change in stock price (convexity term)
$\Theta$ : option theta
$dt$ : passage of time
$\nu$ : option vega
$d\sigma$ : change in implied volatility
$\rho$ : option rho
$dr$ : change in interest rate

The gamma term is particularly important because options are non-linear instruments. For large stock moves, the $(dS)^2$ term can dominate the linear term. Consider a stock that moves $10; the delta contribution is $\Delta \times 10$ , but the gamma contribution is $\frac{1}{2}\Gamma \times 100$ . Even if gamma is much smaller than delta in absolute terms, the square of the price move amplifies its effect significantly.

P&L Attribution ExampleLink Copied

Let's see how this works in practice. First, we define a pricing function to establish the baseline value. Then, we simulate a market move involving simultaneous changes in price, time, and volatility to compare the first-order (Delta, Vega, Theta) and second-order (Gamma) approximations against the actual P&L:

In[18]:

Code

import numpy as np
from scipy.stats import norm


def black_scholes_price(S, K, T, r, sigma, option_type="call"):
    """Calculate Black-Scholes price for European options."""
    if T <= 0:
        if option_type == "call":
            return max(S - K, 0)
        else:
            return max(K - S, 0)

    d1 = (np.log(S / K) + (r + 0.5 * sigma**2) * T) / (sigma * np.sqrt(T))
    d2 = d1 - sigma * np.sqrt(T)

    if option_type == "call":
        return S * norm.cdf(d1) - K * np.exp(-r * T) * norm.cdf(d2)
    else:
        return K * np.exp(-r * T) * norm.cdf(-d2) - S * norm.cdf(-d1)

In[19]:

Code

# Initial parameters
S0 = 100
K = 100
T0 = 0.25  # 3 months
r = 0.05
sigma0 = 0.20

# Calculate initial price and Greeks
price_initial = black_scholes_price(S0, K, T0, r, sigma0, "call")
delta = black_scholes_delta(S0, K, T0, r, sigma0, "call")
gamma = black_scholes_gamma(S0, K, T0, r, sigma0)
theta = black_scholes_theta(S0, K, T0, r, sigma0, "call")
vega = black_scholes_vega(S0, K, T0, r, sigma0)

# Scenario: Stock moves up $2, vol increases 1%, one day passes
dS = 2.0
dT = 1 / 365  # 1 day
dsigma = 0.01

# True new price
S_new = S0 + dS
T_new = T0 - dT
sigma_new = sigma0 + dsigma
price_new = black_scholes_price(S_new, K, T_new, r, sigma_new, "call")
actual_pnl = price_new - price_initial

# First-order approximation
first_order_pnl = delta * dS + theta * dT + vega * dsigma

# Second-order approximation
second_order_pnl = delta * dS + 0.5 * gamma * dS**2 + theta * dT + vega * dsigma

Out[20]:

Console

P&L Attribution Analysis
==================================================

Initial Option Price: $4.6150

Greeks at inception:
  Delta:  0.5695
  Gamma:  0.0393
  Theta:  $-0.0287/day
  Vega:   19.6440

Market Changes:
  Stock: $100.00 → $102.00 (Δ = +$2.00)
  Vol:   20.0% → 21.0% (Δ = +1.0%)
  Time:  1 day passed

P&L Results:
  Actual P&L:              $1.3759
  First-order estimate:    $1.3067  (error: $0.0693)
  Second-order estimate:   $1.3852  (error: $0.0093)

The second-order approximation captures the convexity effect and gives a much more accurate estimate. This is why gamma is needed for risk management; relying only on delta can materially underestimate P&L for larger moves. In this example, the gamma term accounts for the difference between the first-order and second-order estimates, showing the practical need for understanding option curvature.

Out[21]:

Visualization

Horizontal bar chart showing the contribution of each Greek (delta, gamma, theta, vega) to the total P&L, with the actual P&L shown for comparison. — P&L attribution breakdown showing contributions from each Greek. Delta dominates for this small price move, but the gamma contribution improves accuracy. The 'Unexplained' bar shows the unexplained P&L after accounting for all Greeks.

The P&L attribution chart makes clear how each risk factor contributed to the total profit. In this scenario, the stock price increase generated positive delta P&L, while the volatility increase added vega P&L. The theta contribution was slightly negative (time decay), and the gamma contribution captured the non-linearity. The small unexplained residual represents higher-order effects not captured by the second-order Taylor expansion.