nag_pde_bs_1d (d03ncc) (PDF version)
d03 Chapter Contents
d03 Chapter Introduction
NAG Library Manual

NAG Library Function Document

nag_pde_bs_1d (d03ncc)

+ Contents

    1  Purpose
    7  Accuracy

1  Purpose

nag_pde_bs_1d (d03ncc) solves the Black–Scholes equation for financial option pricing using a finite difference scheme.

2  Specification

#include <nag.h>
#include <nagd03.h>
void  nag_pde_bs_1d (Nag_OptionType kopt, double x, Nag_MeshType mesh, Integer ns, double s[], Integer nt, double t[], const Nag_Boolean tdpar[], const double r[], const double q[], const double sigma[], double alpha, Integer ntkeep, double f[], double theta[], double delta[], double gamma[], double lambda[], double rho[], NagError *fail)

3  Description

nag_pde_bs_1d (d03ncc) solves the Black–Scholes equation (see Hull (1989) and Wilmott et al. (1995))
f t +r-qS f S +σ2S22 2f S2 =rf (1)
Smin<S<Smax,  tmin<t<tmax, (2)
for the value f of a European or American, put or call stock option, with exercise price X. In equation (1) t is time, S is the stock price, r is the risk free interest rate, q is the continuous dividend, and σ is the stock volatility. According to the values in the array tdpar, the arguments r, q and σ may each be either constant or functions of time. The function also returns values of various Greeks.
nag_pde_bs_1d (d03ncc) uses a finite difference method with a choice of time-stepping schemes. The method is explicit for alpha=0.0 and implicit for nonzero values of alpha. Second order time accuracy can be obtained by setting alpha=0.5. According to the value of the argument mesh the finite difference mesh may be either uniform, or user-defined in both S and t directions.

4  References

Hull J (1989) Options, Futures and Other Derivative Securities Prentice–Hall
Wilmott P, Howison S and Dewynne J (1995) The Mathematics of Financial Derivatives Cambridge University Press

5  Arguments

1:     koptNag_OptionTypeInput
On entry: specifies the kind of option to be valued.
kopt=Nag_EuropeanCall
A European call option.
kopt=Nag_AmericanCall
An American call option.
kopt=Nag_EuropeanPut
A European put option.
kopt=Nag_AmericanPut
An American put option.
Constraint: kopt=Nag_EuropeanCall, Nag_AmericanCall, Nag_EuropeanPut or Nag_AmericanPut.
2:     xdoubleInput
On entry: the exercise price X.
3:     meshNag_MeshTypeInput
On entry: indicates the type of finite difference mesh to be used:
mesh=Nag_UniformMesh
Uniform mesh.
mesh=Nag_CustomMesh
Custom mesh supplied by you.
Constraint: mesh=Nag_UniformMesh or Nag_CustomMesh.
4:     nsIntegerInput
On entry: the number of stock prices to be used in the finite difference mesh.
Constraint: ns2.
5:     s[ns]doubleInput/Output
On entry: if mesh=Nag_CustomMesh, s[i-1] must contain the ith stock price in the mesh, for i=1,2,,ns. These values should be in increasing order, with s[0]=Smin and s[ns-1]=Smax.
If mesh=Nag_UniformMesh, s[0] must be set to Smin and s[ns-1] to Smax, but s[1],s[2],,s[ns-2] need not be initialized, as they will be set internally by the function in order to define a uniform mesh.
On exit: if mesh=Nag_UniformMesh, the elements of s define a uniform mesh over Smin,Smax.
If mesh=Nag_CustomMesh, the elements of s are unchanged.
Constraints:
  • if mesh=Nag_CustomMesh, s[0]0.0 and s[i-1]<s[i], for i=1,2,,ns-1;
  • if mesh=Nag_UniformMesh, 0.0s[0]<s[ns-1].
6:     ntIntegerInput
On entry: the number of time-steps to be used in the finite difference method.
Constraint: nt2.
7:     t[nt]doubleInput/Output
On entry: if mesh=Nag_CustomMesh then t[j-1] must contain the jth time in the mesh, for j=1,2,,nt. These values should be in increasing order, with t[0]=tmin and t[nt-1]=tmax.
If mesh=Nag_UniformMesh then t[0] must be set to tmin and t[nt-1] to tmax, but t[1],t[2],,t[nt-2] need not be initialized, as they will be set internally by the function in order to define a uniform mesh.
On exit: if mesh=Nag_UniformMesh, the elements of t define a uniform mesh over tmin,tmax.
If mesh=Nag_CustomMesh, the elements of t are unchanged.
Constraints:
  • if mesh=Nag_CustomMesh, t[0]0.0 and t[j-1]<t[j], for j=1,2,,nt-1;
  • if mesh=Nag_UniformMesh, 0.0t[0]<t[nt-1].
8:     tdpar[3]const Nag_BooleanInput
On entry: specifies whether or not various arguments are time-dependent. More precisely, r is time-dependent if tdpar[0]=Nag_TRUE and constant otherwise. Similarly, tdpar[1] specifies whether q is time-dependent and tdpar[2] specifies whether σ is time-dependent.
9:     r[dim]const doubleInput
Note: the dimension, dim, of the array r must be at least
  • nt when tdpar[0]=Nag_TRUE;
  • 1 otherwise.
On entry: if tdpar[0]=Nag_TRUE then r[j-1] must contain the value of the risk-free interest rate rt at the jth time in the mesh, for j=1,2,,nt.
If tdpar[0]=Nag_FALSE then r[0] must contain the constant value of the risk-free interest rate r. The remaining elements need not be set.
10:   q[dim]const doubleInput
Note: the dimension, dim, of the array q must be at least
  • nt when tdpar[1]=Nag_TRUE;
  • 1 otherwise.
On entry: if tdpar[1]=Nag_TRUE then q[j-1] must contain the value of the continuous dividend qt at the jth time in the mesh, for j=1,2,,nt.
If tdpar[1]=Nag_FALSE then q[0] must contain the constant value of the continuous dividend q. The remaining elements need not be set.
11:   sigma[dim]const doubleInput
Note: the dimension, dim, of the array sigma must be at least
  • nt when tdpar[2]=Nag_TRUE;
  • 1 otherwise.
On entry: if tdpar[2]=Nag_TRUE then sigma[j-1] must contain the value of the volatility σt at the jth time in the mesh, for j=1,2,,nt.
If tdpar[2]=Nag_FALSE then sigma[0] must contain the constant value of the volatility σ. The remaining elements need not be set.
12:   alphadoubleInput
On entry: the value of λ to be used in the time-stepping scheme. Typical values include:
alpha=0.0
Explicit forward Euler scheme.
alpha=0.5
Implicit Crank–Nicolson scheme.
alpha=1.0
Implicit backward Euler scheme.
The value 0.5 gives second-order accuracy in time. Values greater than 0.5 give unconditional stability. Since 0.5 is at the limit of unconditional stability this value does not damp oscillations.
Suggested value: alpha=0.55.
Constraint: 0.0alpha1.0.
13:   ntkeepIntegerInput
On entry: the number of solutions to be stored in the time direction. The function calculates the solution backwards from t[nt-1] to t[0] at all times in the mesh. These time solutions and the corresponding Greeks will be stored at times t[i-1], for i=1,2,,ntkeep, in the arrays f, theta, delta, gamma, lambda and rho. Other time solutions will be discarded. To store all time solutions set ntkeep=nt.
Constraint: 1ntkeepnt.
14:   f[ns×ntkeep]doubleOutput
On exit: f[ns×j-1+i-1], for i=1,2,,ns and j=1,2,,ntkeep, contains the value f of the option at the ith mesh point s[i-1] at time t[j-1].
15:   theta[ns×ntkeep]doubleOutput
16:   delta[ns×ntkeep]doubleOutput
17:   gamma[ns×ntkeep]doubleOutput
18:   lambda[ns×ntkeep]doubleOutput
19:   rho[ns×ntkeep]doubleOutput
On exit: the values of various Greeks at the ith mesh point s[i-1] at time t[j-1], as follows:
theta[ns×j-1+i-1]= f t , delta[ns×j-1+i-1]= f S , gamma[ns×j-1+i-1]= 2f S2 , lambda[ns×j-1+i-1]= f σ , rho[ns×j-1+i-1]= f r .
20:   failNagError *Input/Output
The NAG error argument (see Section 3.6 in the Essential Introduction).

6  Error Indicators and Warnings

NE_ALLOC_FAIL
Dynamic memory allocation failed.
NE_BAD_PARAM
On entry, argument value had an illegal value.
NE_INT
On entry, ns=value.
Constraint: ns2.
On entry, nt=value.
Constraint: nt2.
On entry, ntkeep=value.
Constraint: ntkeep1.
NE_INT_2
On entry, ntkeep=value and nt=value.
Constraint: ntkeepnt.
NE_INTERNAL_ERROR
An internal error has occurred in this function. Check the function call and any array sizes. If the call is correct then please contact NAG for assistance.
NE_NOT_STRICTLY_INCREASING
On entry, s[I]s[I-1] in custom mesh: I=value.
On entry, t[J]t[J-1] in custom mesh: J=value.
NE_REAL
On entry, alpha=value.
Constraint: alpha1.0.
On entry, alpha=value.
Constraint: alpha0.0.
On entry, s[0]<0.0: s[0]=value.
On entry, t[0]<0.0: t[0]=value.
NE_REAL_2
On entry, s[ns-1]=value and s[0]=value.
Constraint: s[ns-1]>s[0].
On entry, t[nt-1]=value and t[0]=value.
Constraint: t[nt-1]>t[0].

7  Accuracy

The accuracy of the solution f and the various derivatives returned by the function is dependent on the values of ns and nt supplied, the distribution of the mesh points, and the value of alpha chosen. For most choices of alpha the solution has a truncation error which is second-order accurate in S and first order accurate in t. For alpha=0.5 the truncation error is also second-order accurate in t.
The simplest approach to improving the accuracy is to increase the values of both ns and nt.

8  Parallelism and Performance

nag_pde_bs_1d (d03ncc) is threaded by NAG for parallel execution in multithreaded implementations of the NAG Library.
nag_pde_bs_1d (d03ncc) makes calls to BLAS and/or LAPACK routines, which may be threaded within the vendor library used by this implementation. Consult the documentation for the vendor library for further information.
Please consult the Users' Note for your implementation for any additional implementation-specific information.

9  Further Comments

9.1  Timing

Each time-step requires the construction and solution of a tridiagonal system of linear equations. To calculate each of the derivatives lambda and rho requires a repetition of the entire solution process. The time taken for a call to the function is therefore proportional to ns×nt.

9.2  Algorithmic Details

nag_pde_bs_1d (d03ncc) solves equation (1) using a finite difference method. The solution is computed backwards in time from tmax to tmin using a λ scheme, which is implicit for all nonzero values of λ, and is unconditionally stable for values of λ>0.5. For each time-step a tridiagonal system is constructed and solved to obtain the solution at the earlier time. For the explicit scheme (λ=0) this tridiagonal system degenerates to a diagonal matrix and is solved trivially. For American options the solution at each time-step is inspected to check whether early exercise is beneficial, and amended accordingly.
To compute the arrays lambda and rho, which are derivatives of the stock value f with respect to the problem arguments σ and r respectively, the entire solution process is repeated with perturbed values of these arguments.

10  Example

This example, taken from Hull (1989), solves the one-dimensional Black–Scholes equation for valuation of a 5-month American put option on a non-dividend-paying stock with an exercise price of $50. The risk-free interest rate is 10% per annum, and the stock volatility is 40% per annum.
A fully implicit backward Euler scheme is used, with a mesh of 20 stock price intervals and 10 time intervals.

10.1  Program Text

Program Text (d03ncce.c)

10.2  Program Data

Program Data (d03ncce.d)

10.3  Program Results

Program Results (d03ncce.r)

Produced by GNUPLOT 4.4 patchlevel 0 -40 -20 0 20 40 0 20 40 60 80 100 -10 -5 0 5 10 15 Option Values Derivatives Stock Price Example Program Option Values and Derivatives at 5 Months to Maturity option values q d g l r
Produced by GNUPLOT 4.4 patchlevel 0 -40 -20 0 20 40 0 20 40 60 80 100 -8 -6 -4 -2 0 2 4 6 8 10 12 Option Values Derivatives Stock Price Option Values and Derivatives at 3.5 Months to Maturity option values q d g l r

nag_pde_bs_1d (d03ncc) (PDF version)
d03 Chapter Contents
d03 Chapter Introduction
NAG Library Manual

© The Numerical Algorithms Group Ltd, Oxford, UK. 2014