One-factor short-rate models have a fundamental structural deficiency: every point on the yield curve moves in perfect lockstep, driven by a single source of randomness. Empirical data contradicts this — principal component analysis of yield curve changes consistently shows that two or three factors are needed to explain 85–99% of the variance, corresponding to level, slope, and curvature movements that are far from perfectly correlated. A swaption depends jointly on rates at different tenors, and its correct pricing requires a model that can express the less-than-perfect correlation between, say, the 2-year and 10-year rate.
Multi-factor term structure models are the response. This topic covers the theoretical extension from one factor to two, the canonical tractable implementation (the G2++ two-additive-factor Gaussian model), its mathematical derivation, the general affine consistency conditions that govern which multi-factor models admit closed-form bond prices, and how two-factor models are used in practice for option-adjusted spread computation and butterfly trading.
In this lesson
- Lessons in this topic
- Prerequisites
- Key ideas
The full lesson is for members.
Read the full lessonNew to SelfMadeQuant.com? See membership options or browse the Fixed Income and Interest-Rate Models syllabus.