org.jadira.quant.coremath

Class IntervalBisection

java.lang.Object

org.jadira.quant.coremath.IntervalBisection

All Implemented Interfaces:

CompositeFunction<Double,QuantitativeFunction<Double,Double>>, QuantitativeFunction<Double,QuantitativeFunction<Double,Double>>

public class IntervalBisection
extends Object
implements CompositeFunction<Double,QuantitativeFunction<Double,Double>>

This class can be used to solve the equation f(x) = 0. It will determine the double value of x, where f is a continuous function defined on an interval between [a, b] and f(a) and f(b) have opposite signs. [a, b] are said to bracket the root since there must be at least one root between a and b that yields f(x) = 0. For each iteration, the approach divides the interval in two by determining the midpoint of the interval, c, such that c = (a+b)/2. The method then determines the interval that brackets the root, this being either [a, c] or [c, b]. This is used as the interval for the next iteration. The process is terminated either if the maximum number of iterations is reached, the gap between this result and the previous iteration result is sufficiently small (if configured with IntervalBisection.PrecisionStrategy.BETWEEN_RESULTS) or the value is within the desired distance of root (if configured with IntervalBisection.PrecisionStrategy.TO_INTERVAL. PrecisionStrategy defaults to IntervalBisection.PrecisionStrategy.TO_INTERVAL, the method defaults to 20 iterations, and an accuracy distance of 1e-3 (0.001). Inspired by the example that appears in Java Methods for Financial Engineering (Philip Barker), this implementation includes error handling, a functional programming model and two precision strategies.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	IntervalBisection.PrecisionStrategy

Constructor Summary

Constructors

Constructor and Description
IntervalBisection(double lowerBound, double higherBound)
IntervalBisection(double lowerBound, double higherBound, int iterations) Create a new instance with the given number of iterations and precision
IntervalBisection(double lowerBound, double higherBound, int iterations, double precision, IntervalBisection.PrecisionStrategy precisionStrategy) Create a new instance with the given number of iterations and precision

Method Summary

Modifier and Type	Method and Description
Double	apply(QuantitativeFunction<Double,Double> computeFunction) Apply a function to the input argument T, yielding an appropriate result R.
double	getHigherBound()
int	getIterations()
double	getLowerBound()
double	getPrecision()

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

IntervalBisection

public IntervalBisection(double lowerBound,
                         double higherBound)

IntervalBisection

public IntervalBisection(double lowerBound,
                         double higherBound,
                         int iterations)

Create a new instance with the given number of iterations and precision

Parameters:

lowerBound - The minimum bound for the range to converge from

higherBound - The maximum bound for the range to converge from

iterations - The number of iterations to perform

IntervalBisection

public IntervalBisection(double lowerBound,
                         double higherBound,
                         int iterations,
                         double precision,
                         IntervalBisection.PrecisionStrategy precisionStrategy)

Create a new instance with the given number of iterations and precision

Parameters:

lowerBound - The minimum bound for the range to converge from

higherBound - The maximum bound for the range to converge from

iterations - The number of iterations to perform

precision - The requested precision

precisionStrategy - Indicates approach for managing precision

Method Detail

getLowerBound

public double getLowerBound()

getHigherBound

public double getHigherBound()

getIterations

public int getIterations()

getPrecision

public double getPrecision()

apply

public Double apply(QuantitativeFunction<Double,Double> computeFunction)

Description copied from interface: CompositeFunction

Apply a function to the input argument T, yielding an appropriate result R.

Specified by:

apply in interface CompositeFunction<Double,QuantitativeFunction<Double,Double>>

Specified by:

apply in interface QuantitativeFunction<Double,QuantitativeFunction<Double,Double>>

Parameters:

computeFunction - the input value

Returns:

the result of the function