All Downloads are FREE. Search and download functionalities are using the official Maven repository.

com.opengamma.strata.pricer.impl.tree.TrinomialTree Maven / Gradle / Ivy

The newest version!
/*
 * Copyright (C) 2016 - present by OpenGamma Inc. and the OpenGamma group of companies
 *
 * Please see distribution for license.
 */
package com.opengamma.strata.pricer.impl.tree;

import com.opengamma.strata.basics.value.ValueDerivatives;
import com.opengamma.strata.collect.ArgChecker;
import com.opengamma.strata.collect.array.DoubleArray;
import com.opengamma.strata.pricer.fxopt.RecombiningTrinomialTreeData;

/**
 * Trinomial tree.
 * 

* Option pricing model based on trinomial tree. Trinomial lattice is defined by {@code LatticeSpecification} * and the option to price is specified by {@code OptionFunction}. *

* Option pricing with non-uniform tree is realised by specifying {@code RecombiningTrinomialTreeData}. */ public class TrinomialTree { /** * Price an option under the specified trinomial lattice. *

* It is assumed that the volatility, interest rate and continuous dividend rate are constant * over the lifetime of the option. * * @param function the option * @param lattice the lattice specification * @param spot the spot * @param volatility the volatility * @param interestRate the interest rate * @param dividendRate the dividend rate * @return the option price */ public double optionPrice( OptionFunction function, LatticeSpecification lattice, double spot, double volatility, double interestRate, double dividendRate) { int nSteps = function.getNumberOfSteps(); double timeToExpiry = function.getTimeToExpiry(); double dt = timeToExpiry / (double) nSteps; double discount = Math.exp(-interestRate * dt); DoubleArray params = lattice.getParametersTrinomial(volatility, interestRate - dividendRate, dt); double middleFactor = params.get(1); double downFactor = params.get(2); double upProbability = params.get(3); double midProbability = params.get(4); double downProbability = params.get(5); ArgChecker.isTrue(upProbability > 0d, "upProbability should be greater than 0"); ArgChecker.isTrue(upProbability < 1d, "upProbability should be smaller than 1"); ArgChecker.isTrue(midProbability > 0d, "midProbability should be greater than 0"); ArgChecker.isTrue(midProbability < 1d, "midProbability should be smaller than 1"); ArgChecker.isTrue(downProbability > 0d, "downProbability should be greater than 0"); DoubleArray values = function.getPayoffAtExpiryTrinomial(spot, downFactor, middleFactor); for (int i = nSteps - 1; i > -1; --i) { values = function.getNextOptionValues(discount, upProbability, midProbability, downProbability, values, spot, downFactor, middleFactor, i); } return values.get(0); } /** * Price an option under the specified trinomial tree gird. * * @param function the option * @param data the trinomial tree data * @return the option price */ public double optionPrice( OptionFunction function, RecombiningTrinomialTreeData data) { int nSteps = data.getNumberOfSteps(); ArgChecker.isTrue(nSteps == function.getNumberOfSteps(), "mismatch in number of steps"); DoubleArray values = function.getPayoffAtExpiryTrinomial(data.getStateValueAtLayer(nSteps)); for (int i = nSteps - 1; i > -1; --i) { values = function.getNextOptionValues( data.getDiscountFactorAtLayer(i), data.getProbabilityAtLayer(i), data.getStateValueAtLayer(i), values, i); } return values.get(0); } /** * Compute option price and delta under the specified trinomial tree gird. *

* The delta is the first derivative of the price with respect to spot, and approximated by the data embedded in * the trinomial tree. * * @param function the option * @param data the trinomial tree data * @return the option price and spot delta */ public ValueDerivatives optionPriceAdjoint( OptionFunction function, RecombiningTrinomialTreeData data) { int nSteps = data.getNumberOfSteps(); ArgChecker.isTrue(nSteps == function.getNumberOfSteps(), "mismatch in number of steps"); DoubleArray values = function.getPayoffAtExpiryTrinomial(data.getStateValueAtLayer(nSteps)); double delta = 0d; for (int i = nSteps - 1; i > -1; --i) { values = function.getNextOptionValues( data.getDiscountFactorAtLayer(i), data.getProbabilityAtLayer(i), data.getStateValueAtLayer(i), values, i); if (i == 1) { DoubleArray stateValue = data.getStateValueAtLayer(1); double d1 = (values.get(2) - values.get(1)) / (stateValue.get(2) - stateValue.get(1)); double d2 = (values.get(1) - values.get(0)) / (stateValue.get(1) - stateValue.get(0)); delta = 0.5 * (d1 + d2); } } return ValueDerivatives.of(values.get(0), DoubleArray.of(delta)); } }





© 2015 - 2024 Weber Informatics LLC | Privacy Policy