Skip to content
All library documents

Perpetual Barrier Derivative Pricing with a First-Passage ODE

Article Quant Q&A · Author: chengcj

Summary

The document values a perpetual claim that pays one unit when a non-dividend-paying stock first falls to a lower barrier. It presents two equivalent routes: risk-neutral expectation of the discount factor at the hitting time, or an ordinary differential equation for the value as a function of stock price. The ODE is solved with a power-function trial solution, then contract boundary conditions select the relevant term. For the stated lower-barrier payoff, the resulting value is a power of the ratio between current price and barrier, with the exponent determined by the interest rate and volatility.

The derivation assumes constant rates and volatility, geometric Brownian motion, and no dividends. It refers to a finite-maturity digital option as a route to the perpetual limit, but does not show the probabilistic derivation in detail. The stated boundary condition and formula correspond to a put-like claim; other payoff directions or contract terms require different conditions. The source also contains a likely notation typo in its differential equation, so the displayed solution should be checked against the intended derivatives and boundary setup before reuse.

Key ideas

  • The claim is worth the risk-neutral expected discount factor evaluated at the first time the stock reaches its barrier.
  • A perpetual price can be found by solving a pricing ODE with contract-specific boundary conditions.
  • A power-function trial solution reduces the ODE to a quadratic equation for its exponents.
  • The stated lower-barrier payoff selects the decreasing power solution and gives a value based on the current-price-to-barrier ratio.

Tags

Full text
# Barrier Derivative Pricing


# Barrier Derivative Pricing












Assume constant interest rate $r$ and a stock with current price at $S_0$ that pays no dividend (assume $S_t\ge0$). When the stock price hits the barrier $B$ (where $B<S_0$) you receive \$$1$ and the derivative would terminate. This derivative doesn't have a maturity date. $S_t$ follows geometric Brownian motion with constant volatility $\sigma$.

What is the present value of this derivative?

## Answer by LocalVolatility (score 1, accepted)

https://quant.stackexchange.com/a/32749

As is often the case, there are generally two solution strategies here.

- (Probabilistic) You explicitly solve for the expected discount factor at the first passage time $\nu$ of $S$ to the level $B$ under the risk-neutral probability measure $\mathbb{P}^*$, i.e. \begin{equation} V_0 = \mathbb{E}_{\mathbb{P}^*} \left[ e^{-r \nu} \right]. \end{equation}

- (Differential Equation) The option value $V$ satisfies the ODE \begin{equation} \frac{1}{2} \sigma^2 S^2 \frac{\mathrm{d} V^2}{\mathrm{d} S^2} + r S \frac{\mathrm{d} V}{\mathrm{d} S} - r V = 0 \end{equation} subject to the contract-specific boundary conditions.

I will outline the second approach here and refer to e.g. Chapter 9 in Wilmott (2006) for further details. See e.g. this blog post for a solution to the finite-maturity American digital call option valuation problem using the first approach. In order to obtain the solution for the perpetual case, simply take the limit $T \rightarrow \infty$. The solution to the put is fully analogous.

ODE Approach

The ODE can be rearranged to

\begin{equation} S^2 \frac{\mathrm{d} V^2}{\mathrm{d} S^2} + \lambda S \frac{\mathrm{d} V}{\mathrm{d} S} - \lambda V = 0, \end{equation}

where $\lambda = 2r / \sigma^2$. This equation is of the Euler-Cauchy type and we thus try the solution

\begin{equation} V(S) = S^\beta \end{equation}

and get

\begin{equation} \beta (\beta - 1) S^\beta + \beta \lambda S^\beta - \lambda S^\beta = 0. \end{equation}

This equation holds for all values of $S$ if

\begin{equation} \beta^2 + \beta (\lambda - 1) - \lambda = 0. \end{equation}

Solving for $\beta$ yields

\begin{equation} \beta_\pm = \frac{1}{\sigma^2} \left( -\left( r - \frac{1}{2} \sigma^2 \right) \pm \left( r + \frac{1}{2} \sigma^2 \right) \right) \end{equation}

and we notice that $\beta_+ = 1$ and $\beta_- = -\lambda$. The general solution to the ODE is given by

\begin{equation} V(S) = c_- S^{-\lambda} + c_+ S, \end{equation}

where $c_\pm$ depend on the boundary conditions of the contract. In case of a put option we have the upper boundary condition $\lim_{S \rightarrow \infty} V(S) = 0$, which implied that $c_+ = 0$. The value matching condition at the lower boundary is $U(B) = 1$ and we thus obtain $c_- = B^\lambda$. Consequently,

\begin{equation} V(S) = \left( \frac{S}{B} \right)^{-\lambda}. \end{equation}

References

Wilmott, Paul (2006) Paul Wilmott on Quantitative Finance, Vol. 1: Wiley, 2nd edition.

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.