For a stock that pays no dividends, and has constant interest rates, I know that the relationship can be described with:

$F(t) = s(t). e^{r(T-t)}$

F(t) = Forward price, S(t) = Stock price, r = risk free rate, T = maturity date, t = time now

How does the equation change, when rates are stochastic?


From another textbook, it appears the relationship can be modeled as:

$ F(t) = s(t) / P(t) $

where $ P(t) = e^{-\int_{t}^{T} r(s) ds} $

does this look right?

  • $\begingroup$ You may want to have a look of this question. $\endgroup$ – Gordon Mar 27 '17 at 18:23

The forward price $K$, determined at time $t$, is the amount such that the payoff at time $T$ is $S_T-K$, while the value at time $t$ is zero. That is, \begin{align*} B_t E\left(\frac{S_T-K}{B_T}\mid \mathcal{F}_t \right)= 0, \end{align*} Where $E$ is the risk-neutral expectation operator. Then, \begin{align*} K&=\frac{E\left(\frac{S_T}{B_T}\mid \mathcal{F}_t \right)}{E\left(\frac{1}{B_T}\mid \mathcal{F}_t \right)}\\ &=\frac{E\left(\frac{S_T}{B_T}\mid \mathcal{F}_t \right)}{\frac{1}{B_t}E\left(\frac{B_t}{B_T}\mid \mathcal{F}_t \right)}\\ &=\frac {\frac{ S_t}{ B_t} }{\frac{1}{B_t}P (t,T)}\\ &=\frac{S_t}{P(t,T)}, \end{align*} where \begin{align*} P (t,T) &= E\left(\frac{B_t}{ B_T}\mid \mathcal {F}_t \right)\\ &=E\left(e^{-\int_t^T r_s ds}\mid \mathcal {F}_t \right) \end{align*} is the price at time $t$ of a zero-coupon bond with maturity $T$ and unit face value.

Alternatively, at time $t$,

  • enter into a forward contract with forward price $K$, which has zero cost at time $t$,

  • short one share with income $S_t$, and

  • long $\frac{S_t}{P (t,T)}$ units of zero-coupon bond with maturity $T$.

The net cost at time $t$ is zero. At maturity $T$,

  • the forward contract has value $S_T-K$,

  • the short position of one share has value $-S_T$, and

  • the zero-coupon bond has value $\frac{S_t}{P (t,T)}$.

Assuming arbitrage free, the value at time $T$ is then \begin{align*} S_T-K - S_T + \frac{S_t}{P (t,T)} = \frac{S_t}{P (t,T)}-K =0, \end{align*} that is, \begin{align*} K=\frac{S_t}{P (t,T)}. \end{align*}

  • $\begingroup$ Thanks Gordon, that is a great answer. Please can you confirm that $P(t,T)$ is the price at time t of a zero coupon bond that pays 1 dollar at time T, i.e $$P(t,T) = e^{ - \int_{t}^{T} r(s) ds}$$ and $B_t$ is the time value of the money market account i.e $$B_t = e^{ \int_{0}^{t} r(s) ds}$$ Finally, in this derivation, are you assuming interest rates to be deterministic? $\endgroup$ – Calypso Mar 28 '17 at 6:46
  • $\begingroup$ I don't assume deternistic interest rate. $P (t,T)$ should be the conditioanl expectation. $B_t$ is correct. $\endgroup$ – Gordon Mar 28 '17 at 12:48
  • $\begingroup$ Thanks Gordon, to summarise then: $$ $$ $P(t,T)$ is the expected (using risk neutral measure) price at time t of a zero coupon bond that pays 1 dollar at time T, i.e $$ P(t,T) = E_Q [ e^{ - \int_{t}^{T} r(s) ds} ] $$ $B_t$ is the time value of the money market account i.e $$ B_t = e^{ \int_{0}^{t} r(s) ds} $$ Assumptions:$$ $$ 1. Stock price and the interest rate are independent. 2. The market is frictionless, arbitrage free, infinite liquidity and complete $\endgroup$ – Calypso Mar 28 '17 at 20:13
  • 1
    $\begingroup$ The independence assumption is not needed. $\endgroup$ – Gordon Mar 30 '17 at 3:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.