= Solution
An <interest rate> model must describe a whole <term structure of interest rates>, not just one future scalar rate. Let $P(t,T)$ be the time-$t$ price of a unit <zero-coupon bond> maturing at $T$. For a sufficiently regular maturity curve, define the <instantaneous forward rate> and <short rate> by
$$
f(t,T)=-\partial_T\log P(t,T),\qquad r_t=f(t,t).
$$
Since $P(t,t)=1$, integration gives
$$
P(t,T)=\exp\left(-\int_t^T f(t,u)\,du\right),\qquad
B_t=\exp\left(\int_0^t r_s\,ds\right).
$$
Here $B$ is the <continuous-time bank account>. Thus modelling the two-parameter field $f(t,T)$ determines both the bonds and the discounting account. A <Gaussian random field> means that every finite collection of its values is jointly Gaussian; merely making each individual forward rate normally distributed is insufficient.
A useful arbitrage-free construction works under a money-market <risk-neutral measure> $Q$. Take independent <Brownian motions> $W^j$, deterministic coefficient sequences $\sigma_j(t,T)$, and a deterministic initial forward curve $f(0,T)$. With the coefficients viewed as vectors in a separable <Hilbert space> $H$, write
$$
df(t,T)=\alpha(t,T)dt+\langle\sigma(t,T),dW_t\rangle_H,\qquad0\leq t\leq T.
$$
For infinitely many factors, the last term means the $L^2$ limit of $\sum_j\sigma_j(t,T)dW_t^j$, not an assumption that the infinite vector of <Brownian motions> is itself $H$-valued. Require $\int_0^t\|\sigma(s,T)\|_H^2ds<\infty$ and the maturity/time regularity needed for stochastic integration and exchange of integrals. Deterministic <drift> and volatility make this a <Gaussian forward-rate field>.
The field <covariance> is determined by
$$
c_t(T,U)=\langle\sigma(t,T),\sigma(t,U)\rangle_H,\qquad
\operatorname{Cov}(f(t,T),f(s,U))=\int_0^{\min(t,s)}c_v(T,U)\,dv.
$$
For every finite set of maturities, $(c_t(T_i,T_j))$ must be <positive semidefinite>. A one-factor model has rank-one instantaneous <covariance>, while additional or infinitely many factors permit richer correlations along the <yield curve>. Equivalently one can drive the curve with a maturity-correlated Brownian field $Z_t(T)$ satisfying $d\langle Z(T),Z(U)\rangle_t=c(T,U)dt$. A <Brownian sheet> gives the example
$$
\operatorname{Cov}(Z_t(T),Z_s(U))=\min(t,s)\min(T,U).
$$
Multiplication by deterministic maturity-dependent amplitudes, or deterministic increasing changes of the two sheet coordinates, gives further Gaussian <covariance> structures. A <covariance> specification alone is not sufficient: the forward <drift> must also obey absence of <arbitrage>.
To derive that <drift> restriction, put
$$
\Sigma(t,T)=\int_t^T\sigma(t,u)\,du.
$$
Differentiate the moving lower endpoint in $\log P(t,T)=-\int_t^T f(t,u)du$ to obtain
$$
d\log P(t,T)=\left(r_t-\int_t^T\alpha(t,u)du\right)dt-\langle\Sigma(t,T),dW_t\rangle_H.
$$
The <Itô formula> therefore gives
$$
\frac{dP(t,T)}{P(t,T)}
=\left(r_t-\int_t^T\alpha(t,u)du+\tfrac12\|\Sigma(t,T)\|_H^2\right)dt
-\langle\Sigma(t,T),dW_t\rangle_H.
$$
Discounted bond prices must be <martingales> under $Q$. Their <drift> vanishes precisely when $\int_t^T\alpha(t,u)du=\tfrac12\|\Sigma(t,T)\|_H^2$. Differentiating in maturity gives the <Heath-Jarrow-Morton model> restriction
$$
\boxed{\alpha(t,T)=\langle\sigma(t,T),\Sigma(t,T)\rangle_H
=\int_t^T c_t(T,u)\,du.}
$$
For scalar amplitudes $v(t,T)$ multiplying a field with <covariance> kernel $c(T,U)$, this becomes $\alpha(t,T)=v(t,T)\int_t^T v(t,u)c(T,u)du$. The extra maturity correlation factor cannot in general be dropped.
The complete Gaussian specification is consequently
$$
f(t,T)=f(0,T)+\int_0^t\int_s^T c_s(T,u)\,du\,ds
+\int_0^t\langle\sigma(s,T),dW_s\rangle_H.
$$
Its mean and <covariance> follow immediately from deterministic integration and the <Itô isometry>. The initial <yield curve> is an input, so fitting initial bond prices requires choosing $f(0,T)$ rather than changing a few scalar parameters. With the displayed <drift>, the discounted bond obeys
$$
\frac{d(P(t,T)/B_t)}{P(t,T)/B_t}=-\langle\Sigma(t,T),dW_t\rangle_H.
$$
On finite horizons, deterministic <square-integrable> bond volatility gives an expectation-one <stochastic exponential>, hence a true <discounted bond price martingale>. This supplies an arbitrage-free pricing model for admissible <portfolios> in any finite collection of these bonds. Under a physical measure, a <market price of risk> vector changes the <drift>: if $dW_t^Q=dW_t^P+\lambda_tdt$, then the physical forward <drift> is $\alpha(t,T)+\langle\sigma(t,T),\lambda_t\rangle_H$. Historical mean changes must therefore not be inserted in place of the risk-neutral <drift> restriction. Gaussianity under both measures additionally requires suitable deterministic <drift> changes; state-dependent risk premiums need not preserve Gaussianity under the physical measure.
There is also a useful general consistency test for a proposed Gaussian field beyond this independent-increment construction. Define
$$
L_t^T=\int_0^t f(s,s)\,ds+\int_t^T f(t,u)\,du,
\qquad B_t^{-1}P(t,T)=e^{-L_t^T}.
$$
In the filtration generated by a Gaussian field, conditional distributions of such linear functionals are Gaussian. For $s\leq t\leq T$, the <martingale> requirement is therefore exactly
$$
\boxed{\mathbb E_Q[L_t^T\mid\mathcal F_s]-\tfrac12\operatorname{Var}_Q(L_t^T\mid\mathcal F_s)=L_s^T.}
$$
This follows by taking the conditional exponential moment and equating it to $e^{-L_s^T}$. It constrains both the deterministic mean and the covariance-induced conditional mean. Matching only unconditional discounted bond <expectations> does not suffice. For the Brownian-field construction, the <Heath-Jarrow-Morton model> restriction realizes this consistency condition through local <drift> cancellation.
Gaussian structure makes <risk-neutral pricing> particularly tractable. As a finite-factor example, take the <Vasicek model> under $Q$:
$$
dr_t=a(b-r_t)dt+\eta dW_t^Q,\qquad a>0.
$$
Solving this <Ornstein-Uhlenbeck process> gives, for $u\geq t$,
$$
r_u=b+(r_t-b)e^{-a(u-t)}+\eta\int_t^u e^{-a(u-v)}dW_v^Q.
$$
Hence, with $\tau=T-t$ and $B(\tau)=(1-e^{-a\tau})/a$, the conditional Gaussian integral $I=\int_t^T r_u du$ has
$$
\mathbb E[I\mid\mathcal F_t]=b\tau+(r_t-b)B(\tau),
\qquad
\operatorname{Var}(I\mid\mathcal F_t)=\frac{\eta^2}{a^2}\left[\tau-2B(\tau)+\frac{1-e^{-2a\tau}}{2a}\right].
$$
The <exponential moment> identity for a normal variable now gives the <zero-coupon bond> price $P(t,T)=\mathbb E_Q[e^{-I}\mid\mathcal F_t]=A(\tau)e^{-B(\tau)r_t}$, where
$$
\log A(\tau)=\left(b-\frac{\eta^2}{2a^2}\right)(B(\tau)-\tau)-\frac{\eta^2}{4a}B(\tau)^2.
$$
This outlines the derivation of the familiar exponential-affine form without assuming a deterministic future discount rate. More general forward-field models need not reduce to a finite-dimensional Markov short-rate process.
For another application, a call at time $T$ on a unit maturity-$U$ bond, with $t<T<U$ and strike $K>0$, is priced under the <T-forward measure>, using $P(t,T)$ as <numéraire>. The bond-price ratio $F_s=P(s,U)/P(s,T)$ is a <martingale> under that measure and has deterministic volatility $-(\Sigma(s,U)-\Sigma(s,T))$. It is consequently lognormal, with integrated log-variance
$$
V=\int_t^T\|\Sigma(s,U)-\Sigma(s,T)\|_H^2ds.
$$
Integrating the lognormal payoff gives the <Gaussian bond-option formula>
$$
\boxed{C_t=P(t,U)\Phi(d_1)-KP(t,T)\Phi(d_2),\quad
d_1=\frac{\log(P(t,U)/(KP(t,T)))+V/2}{\sqrt V},\quad d_2=d_1-\sqrt V.}
$$
At $V=0$ use $(P(t,U)-KP(t,T))^+$. The same idea prices other suitable Gaussian-rate derivatives through forward measures and normal exponential moments.
The attractions are exact initial-curve fitting, explicit <covariance> control across maturities, and tractable bond and option prices. The limitations are also structural: Gaussian <short rates> and <instantaneous forward rates> generally can be negative; finite-factor specifications restrict maturity correlations; and deterministic volatilities cannot reproduce arbitrary stochastic changes in rate dispersion. An infinite-factor model allows richer <covariance> but typically leaves risks that cannot be spanned by finitely many traded bonds. Thus Gaussianity is a useful modelling choice, not by itself a guarantee of positivity, <market completeness>, or absence of <arbitrage>.
Back to article page
