Hull-White Model — An Extended Vasicek Model with Time-Varying Parameters

The Hull-White model is a classic framework in the field of interest rate modeling, proposed by John Hull and Alan White in 1990. It extends the Vasicek model by introducing a time-varying parameter θ(t), enabling the model to precisely fit the market's initial yield curve while retaining mean-reversion characteristics.

🔷 Stochastic Differential Equation (SDE)

🔶 Expression for the Time-Varying Parameter θ(t)

r(t) Short-term interest rate a Mean reversion speed σ Volatility θ(t) Time-varying mean reversion level W(t) Standard Brownian motion f(0,t) Initial instantaneous forward rate

Interest Rate Path Simulation

Model Parameters
θ(t) Time-Varying Parameter  θ(t)=θL+(θS-θL)e-κt
Simulation Settings
—
Mean Final Rate
—
Standard Deviation of Final Rate
—
Minimum Final Rate
—
Maximum Final Rate
—
Terminal θ(T)/a

Deeply Understanding the Hull-White Model

The Hull-White model is one of the most important interest rate models in financial engineering. The following helps you comprehensively understand the value and limitations of this model from multiple dimensions.

Standard Introduction

The Hull-White model is an extension of the Vasicek model, proposed by John Hull and Alan White in 1990. Its core feature is the time-varying parameter θ(t), which enables the model to precisely fit the market's initial yield curve. The model belongs to the class of single-factor no-arbitrage interest rate models, in which the short rate r(t) follows a mean-reverting process, fluctuating around a time-varying level. This model is widely used in interest rate derivatives pricing, risk management, and asset-liability management.

Plain-Language Introduction

Imagine the interest rate as a small ball pulled by a rubber band—the anchor point of the rubber band moves over time. Interest rates will not rise or fall indefinitely, but are "pulled back" toward a changing "normal level." This normal level θ(t)/a is determined by market data and evolves over time. The volatility σ controls the "jitter" magnitude of the interest rate, and the mean reversion speed a controls the "pull" strength of the rubber band.

What Problem Does It Solve?

  • Fitting the market yield curve: Through the time-varying θ(t), it precisely matches any initial term structure, overcoming the defect that the Vasicek model cannot fit the market curve.
  • Interest rate derivatives pricing: It provides analytical bond price formulas and European option pricing formulas (based on the normal distribution).
  • Mean reversion modeling: It captures the economic reality that interest rates do not diverge indefinitely.

Why Is It Important?

  • Bridge between theory and practice: It combines mathematical tractability with market consistency and is a standard tool in both academia and industry.
  • Regulatory recognition: It is widely used in bank interest rate risk management (IRRBB) and solvency assessment.
  • Flexible extension: It can be extended to multi-factor models (such as the Hull-White two-factor model) to better fit the volatility surface.

Application Scenarios

  • Interest rate derivatives pricing: Interest rate caps/floors, swaptions, Bermudan options, etc.
  • Asset-liability management: Banks and insurance companies simulate future interest rate scenarios and assess asset-liability matching.
  • Mortgage valuation: Pricing of mortgage-backed securities (MBS) with prepayment options.
  • Risk measurement: Interest rate scenario generation in the calculation of VaR, ES, and other risk metrics.

Model Deficiencies

  • Single-factor limitation: All interest rates are perfectly correlated and cannot capture twists and butterfly shifts in the yield curve.
  • Normal distribution assumption: Interest rates can become negative (although this may be seen as an advantage in a low interest rate environment).
  • Constant volatility: It cannot fit the market volatility smile/skew and is inaccurate for pricing deep out-of-the-money options.
  • Calibration instability: θ(t) is sensitive to market data, and intraday changes may be large.