I've been working with a convexity adjustment for an interest rate payoff and the next question came to me:

The usual problem that gives rise to the convexity adjustment I'm referring to is as follows:

Consider a series of dates,

\begin{align*} 0 \leq t_0 \leq T_s < T_p < T_e, \end{align*}

Where $T_p$ is the time of payment, $T_s$ is the time when the interest rate is observed and $T_e$ is the time where the interest of the forward rate ends.

If a payment was to ve priced in time $T_p$, one would calculate the expectation over the measure $Q_{T_p}$, as follows:

\begin{align*} P(t_0, T_p)E^{T_p}\big(L(T_s, T_s, T_e) \mid \mathcal{F}_{t_0}\big). \end{align*}

Now changing from the measure $Q_{T_p}$ to the natural measure $Q_{T_e}$ would induce an extra term in the expectation that would lead us to the convexity adjustment:

\begin{align*} &\ P(t_0, T_p)E^{T_p}\big(L(T_s, T_s, T_e) \mid \mathcal{F}_{t_0}\big) \\ =&\ P(t_0, T_p)E^{T_e}\Big(\frac{dQ_{T_p}}{dQ_{t_e}}L(T_s, T_s, T_e) \mid \mathcal{F}_{t_0}\Big)\\ =&\ P(t_0, T_p)E^{T_e}\Big(\frac{P(t_0, T_e)}{P(t_0, T_p)P(T_p, T_e)} L(T_s, T_s, T_e)\mid \mathcal{F}_{t_0}\Big)\\ =&\ P(t_0, T_e)E^{T_e}\Big(\frac{1}{P(T_p, T_e)} L(T_s, T_s, T_e)\mid \mathcal{F}_{t_0}\Big)\\ =&\ P(t_0, T_e)E^{T_e}\Big(\big(1+ \Delta_p^e L(T_p, T_p, T_e) \big) L(T_s, T_s, T_e)\mid \mathcal{F}_{t_0}\Big)\\ =&\ P(t_0, T_e)E^{T_e}\Big(\big(1+ \Delta_p^e L(T_s, T_p, T_e) \big) L(T_s, T_s, T_e)\mid \mathcal{F}_{t_0}\Big) \end{align*}

Now, what if the date sequence was like this:

\begin{align*} 0 \leq t_0 \leq T_s < T_e < T_p, \end{align*}

With the payment made in $T_p$ after the interest rate term ends. Would a change of measure and an adjustment be needed to price the payoff? I'd think that if that's the case it would be something like this:

\begin{align*} &\ P(t_0, T_p)E^{T_p}\big(L(T_s, T_s, T_e) \mid \mathcal{F}_{t_0}\big) \\ =&\ P(t_0, T_e)E^{T_e}\Big(P(T_e, T_p) L(T_s, T_s, T_e)\mid \mathcal{F}_{t_0}\Big)\\ =&\ P(t_0, T_e)E^{T_e}\Big(\frac{1}{1+ \Delta_e^p L(T_e, T_e, T_p)} L(T_s, T_s, T_e)\mid \mathcal{F}_{t_0}\Big)\\ \end{align*}

But I don't see how a change of measure with the same logic as the former case would induce this term. On the other hand, it might be possible that since the payment is done after $T_e$ it doesn't make much sense to make the adjustment since the interest period is over (in theory) and one would price the payoff up to $T_e$ and maybe discount it from $T_p$.

I hope I made myself clear,

Much help appreciated


A convexity adjustment arises regardless of $T_p$ being before or after $T_e$: $$ E^{T_p}\left[L(T_s, T_s, T_e) \right] = E^{T_e}\left[\frac{dQ^{T_p}}{dQ^{T_e}}L(T_s, T_s, T_e) \right] = E^{T_e}\left[\frac{P(t_0, T_e) P(T_s,T_p)}{P(t_0, T_p)P(T_s,T_e)}L(T_s, T_s, T_e) \right] $$ If $T_p < T_e$ then $\frac{P(T_s,T_p)}{P(T_s,T_e)} = 1 + \Delta_p^e L(T_s, T_p, T_e)$ and you get your original formula.

If $T_p > T_e$ then $\frac{P(T_s,T_p)}{P(T_s,T_e)} = \frac{1}{1 + \Delta_e^p L(T_s, T_e, T_p)}$ and you get the formula you were looking for.


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.