Libor Market Model
LIBOR Market Model (Brace-Gatarek-Musiela) · Also known as: BGM Model, LMM
The LIBOR Market Model (BGM), developed by Brace, Gatarek, and Musiela (1997), is a multi-factor interest rate model that directly models forward LIBOR rates as lognormal processes. Unlike short-rate models, LMM naturally prices caplets at the market level and is the industry standard for valuing caps, floors, and exotic interest rate derivatives.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Use LMM when pricing caps, floors, and swaptions with multiple reset dates. It is essential for exotic interest rate products tied to LIBOR. LMM is the market standard for swaption pricing. However, it can be slow for path-dependent products and may require approximate measures (terminal, spot) to reduce simulation burden.
Strengths & limitations
- Direct market instruments: models LIBOR rates directly, making calibration to caps and floors straightforward
- Lognormal dynamics: ensures positive rates naturally; consistent with Black caplet model
- Closed-form caplet prices: under the standard measure, caplets price exactly to the Black formula
- Industry standard: widely adopted in investment banks for consistent pricing across interest rate books
- High dimensionality: simulating many LIBOR rates (one for each tenor) is computationally expensive
- Correlation specification: the correlation matrix between LIBOR rates is not uniquely determined; over-parameterization risk
- Caplet smile: standard LMM cannot reproduce caplet implied volatility smiles without extensions (stochastic vol, jumps)
- Drift complexity: the drift depends on the choice of numéraire; errors in measure change lead to pricing artifacts
Frequently asked
How does LMM differ from short-rate models?
Short-rate models (Hull-White, Vasicek) model an abstract instantaneous rate r(t) and derive LIBOR as a function of the curve. LMM models LIBOR directly as the fundamental variable. LMM has the advantage that caplets price analytically to the Black formula; short-rate models must numerically solve for caplet prices.
What is a numéraire in LMM?
A numéraire is a choice of discount factor used to define the risk-neutral measure. In LMM, different choices simplify different products: the terminal measure prices options at the longest tenor; the spot measure prices bonds at all dates. Changing numéraires requires Radon-Nikodym derivatives, which become drift adjustments in the SDE.
How do I handle the curse of dimensionality in LMM?
Reduce dimensionality by using a small number of principal components of the correlation matrix (e.g., 2-3 instead of N). Use low-discrepancy (quasi-Monte Carlo) sampling to reduce paths needed per factor. Approximate measures (terminal, spot) speed up simulation by avoiding full path evolution.
Why does standard LMM not produce caplet smile?
LMM assumes LIBOR rates are lognormal; Black's model prices caplets under the lognormal assumption. If the true caplet implied volatility exhibits smile (varies with strike), standard LMM cannot reproduce it. Extensions using stochastic volatility (SABR applied to LIBOR) or jump-diffusion add smile.
Sources
- Brace, A., Gatarek, D., & Musiela, M. (1997). The market model of interest rate dynamics. Mathematical Finance, 7(2), 127-155. DOI: 10.1111/1467-9965.00028 ↗
- Jamshidian, F. (1997). LIBOR and swap market models and measures. Finance and Stochastics, 1(4), 293-330. DOI: 10.1007/s007800050026 ↗
How to cite this page
ScholarGate. (2026, June 3). LIBOR Market Model (Brace-Gatarek-Musiela). ScholarGate. https://scholargate.app/en/quantitative-finance/libor-market-model
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- Change of NumeraireQuantitative Finance↔ compare
- HJM FrameworkQuantitative Finance↔ compare
- Hull-White ModelQuantitative Finance↔ compare
- Risk-Neutral ValuationQuantitative Finance↔ compare