Option Pricing Model
An option pricing model is a mathematical framework that calculates the theoretical fair value of an option contract based on the characteristics of the underlying asset and the option's terms. The most widely known is the Black-Scholes-Merton model (1973); others include the binomial tree model, the Cox-Ross-Rubinstein model, and stochastic volatility models such as Heston.
Key takeaways
- Black-Scholes assumes constant volatility, log-normal returns, no dividends, and continuous trading — simplifications that limit real-world accuracy.
- The binomial tree model discretizes time into steps, allowing valuation of American options and accommodating changing volatility.
- Stochastic volatility models (Heston, SABR) treat volatility as a random process, better capturing the volatility smile observed in markets.
- Implied volatility is the market-derived volatility input that makes a model price equal to the observed market price.
- Local volatility models (Dupire) produce an exact fit to the entire volatility surface but have poor forward-volatility dynamics.
Explanation
Option pricing models resolve one of the fundamental challenges in derivatives markets: determining a rational, arbitrage-free price for an instrument whose value depends on an uncertain future outcome. Before Black and Scholes (1973), option pricing relied on intuition and market convention. Black, Scholes, and Merton's insight was that in a world where continuous delta-hedging is possible, a risk-free portfolio can be constructed from an option and its underlying, allowing the option price to be derived through a no-arbitrage argument rather than subjective probability forecasting.
The Black-Scholes-Merton (BSM) model derives its option pricing formula from the assumption that the underlying asset follows geometric Brownian motion with constant drift and volatility. Under these assumptions, the call price is: C = S × N(d₁) − K × e^(−rT) × N(d₂), where d₁ = [ln(S/K) + (r + σ²/2)T] / (σ√T) and d₂ = d₁ − σ√T, and N(·) is the cumulative normal distribution function. The formula is elegant, computationally instant, and Nobel Prize-winning — but its assumptions are frequently violated in practice.
The binomial tree model (Cox-Ross-Rubinstein, 1979) discretizes the option's life into time steps and computes the option value recursively backward from expiry, where the payoff is known. At each node, the underlying can move up by factor u or down by factor d. Risk-neutral probabilities are assigned to each move, ensuring the model is consistent with the risk-free rate. The binomial model handles American options naturally by comparing the intrinsic value (immediate exercise) with the continuation value at each node, and it converges to the Black-Scholes price as the number of time steps increases.
The persistent observation of the 'volatility smile' — where implied volatility is higher for out-of-the-money and in-the-money options than for at-the-money options — reveals that the constant-volatility assumption of BSM is empirically invalid. Markets price OTM puts with higher implied volatility to reflect fat-tail risk (crash protection premium) and OTM calls with higher implied vol to reflect jump risk. Stochastic volatility models (Heston model, SABR model) address this by modeling volatility as a mean-reverting random process that is correlated with the underlying asset return — capturing the 'skew' observed in equity options markets, where falling prices are typically accompanied by rising volatility.
For exotic options — barrier options, Asian options, quanto options — closed-form solutions often do not exist, and numerical methods (Monte Carlo simulation, finite difference PDE solvers) must be used. The model choice matters: different models can produce meaningfully different prices for path-dependent options, and calibration to the market-observed volatility surface is critical to ensure consistency with liquid vanilla options before pricing exotic structures.
Formula
BSM Call: C = S·N(d₁) − K·e^(−rT)·N(d₂); d₁ = [ln(S/K) + (r + σ²/2)T] / (σ√T); d₂ = d₁ − σ√T
Example
A dealer quotes a 3-month at-the-money call option on a European equity index (current level 4,000) with the following inputs: K = 4,000, T = 0.25 years, r = 4.0%, σ = 20% (implied vol), no dividends. Using Black-Scholes: d₁ = [ln(1) + (0.04 + 0.02) × 0.25] / (0.20 × 0.50) = [0 + 0.015] / 0.10 = 0.15; d₂ = 0.15 − 0.10 = 0.05; N(0.15) ≈ 0.5596; N(0.05) ≈ 0.5199. Call price = 4000 × 0.5596 − 4000 × e^(−0.04×0.25) × 0.5199 = 2,238.4 − 4000 × 0.9900 × 0.5199 = 2,238.4 − 2,058.8 = $179.6 per unit. The delta of the call is N(d₁) = 0.56, meaning the dealer immediately hedges by buying 56% of the notional index value. If the market subsequently prices the same option at $190 (due to a vol spike), the implied vol has risen from 20% to approximately 21.5%.
Related terms
Arbitrage At The Money Binomial Tree Model Brownian Motion Call Option Covered Call Cox Ross Rubinstein Model Delta Equity Equity Index Exotic Options Expiration Date