Risk-Neutral Pricing · Expectation Under Measure Change
Risk-Neutral Pricing — Expectation Under Measure Change
Core Formulas
Mathematical FoundationInteractive Simulation
Adjust parameters to observe changesUnderstanding Risk-Neutral Pricing
📘 Standard Introduction
Risk-Neutral Pricing is the core methodology of financial engineering. In a risk-neutral world (measure Q), all investors are risk-neutral, and the expected return on assets equals the risk-free rate r. The current value of any derivative equals the risk-free discount of its expected payoff under the risk-neutral measure: V₀ = e⁻ʳᵀ 𝔼^Q[V_T]. The measure change is achieved through Girsanov's theorem, where the Radon-Nikodym derivative dQ/dP adjusts the Brownian motion drift under the real measure P from μ to r.
💬 Plain-Language Introduction
Imagine two parallel worlds: in the real world (P), stocks rise by an average of 10%, but with high risk; in the "neutral world" (Q), stocks rise only by 5% (the risk-free rate), but everyone here ignores risk. Miraculously, in both worlds, the correct price of the option is the same! We choose to calculate in the neutral world—because the mathematics is simpler there, and we do not need to worry about what the "risk compensation" should be. The core technique is to use a mathematical tool called "measure change" to "shift" the real-world distribution into the neutral world.
🎯 What Problem Does It Solve?
In the real world, different assets have different risk premiums, and pricing derivatives requires knowing the "correct discount rate"—which is nearly impossible. Risk-neutral pricing eliminates the risk preference parameter: the option price depends only on the risk-free rate r and volatility σ, and is independent of the real drift rate μ. This made the Black-Scholes formula possible and laid the foundation for the entire theory of derivatives pricing.
⭐ Why Is It Important?
① No-arbitrage pricing: guarantees that prices contain no arbitrage opportunities; ② Unified framework: all derivatives (options, futures, swaps, etc.) are priced using the same method; ③ Basis for numerical computation: Monte Carlo simulation, tree methods, and PDE methods all rely on the risk-neutral measure; ④ Independent of reality: pricing does not depend on the unobservable expected return μ but only on observable market parameters.
🏦 Application Scenarios
- Option market making: exchanges and market makers calculate theoretical option prices in real time
- Structured products: banks price structured products such as principal-protected wealth management products and snowball products when designing them
- Risk management: calculate VaR and the Greeks for risk hedging
- Corporate valuation: fair value assessment of convertible bonds, employee options, and real options
- Fixed income: pricing of interest rate derivatives and credit derivatives
⚠️ Limitations and Deficiencies
- Complete market assumption: requires frictionless markets, continuous trading, and no transaction costs—which do not exist in reality
- Constant volatility assumption: the BS model assumes σ is constant, but actual volatility changes (there is a "volatility smile")
- Continuous diffusion process: cannot handle jumps in asset prices (such as price gaps caused by sudden news)
- Constant interest rate: extended models require a stochastic interest rate framework
- Liquidity risk: does not consider extreme situations of drying market liquidity