Risk Latte - Installment Warrant Valuation using Gaussian Quadrature Methods

Installment Warrant Valuation using Gaussian Quadrature Methods

Team Latte
March 8, 2011

Here is a very interesting and a possibly a very attractive retail product. Letís say thatDelhi Bombay bank issues installment put warrant on Hang Seng Index (HSI). The warrants have the following structure: the investor pays $3.00 on the purchase date. One year later, the investor may choose to investment another $3.00 and receive an American put option on the Hang Seng Index, making the total cost of the warrant as $6.00. Alternative, he may choose not to pay the $3.00 and let the option expire worthless.

How do we value this installment warrant (compound option)?

There are quite a few numerical routines to evaluate this product. For example, if the underlying option and the compound option are both of European style then Rubenstein has suggested a Newton Raphson method to solve a non-linear equation. However, if the compound option becomes American in style then such a method does not yield a robust result.

Gaussian Quadrature (GQ) methods can be used to estimate the value of the above American style compound option in a robust way. The value of the above option would be given by:

In the above, and are strike prices and is the underlying put option and is the normal density function.

Gaussian Quadrature methods are a robust way of estimating numerical integrals. Essentially, we approximate the integrable function by the zeros of the orthogonal polynomials which are integration points (the abscissas) on the plane and , the weighting function related to the orthogonal polynomials.

There are many kinds of Gaussian Quadrature (GQ) algorithms. Here are three important Gaussian quadrature algorithms.

  1. Gauss-Kronrod local quadrature method (an adaptive routine of Q1DA by Kahaner, Moler and Nash is based on Gauss-Konrod method and is suggested by IzzyNelken to value an installment warrant like the one mentioned above).

  2. Gauss-Legendre quadrature method (we explain and use this method to solve valuation problems in our CFE classes);

  3. Gauss-Laguerre quadrature method.

Any comments and queries can be sent through our web-based form.

More on Quantitative Finance >>

back to top


More from Articles

Quantitative Finance