GJRGARCHProcess Class Reference
[Stochastic processes]

Stochastic-volatility GJR-GARCH(1,1) process. More...

#include <ql/processes/gjrgarchprocess.hpp>

Inheritance diagram for GJRGARCHProcess:

List of all members.

Public Types

enum  Discretization { PartialTruncation, FullTruncation, Reflection }

Public Member Functions

 GJRGARCHProcess (const Handle< YieldTermStructure > &riskFreeRate, const Handle< YieldTermStructure > &dividendYield, const Handle< Quote > &s0, Real v0, Real omega, Real alpha, Real beta, Real gamma, Real lambda, Real daysPerYear=252.0, Discretization d=FullTruncation)
Size size () const
 returns the number of dimensions of the stochastic process
Disposable< ArrayinitialValues () const
 returns the initial values of the state variables
Disposable< Arraydrift (Time t, const Array &x) const
 returns the drift part of the equation, i.e., $ \mu(t, \mathrm{x}_t) $
Disposable< Matrixdiffusion (Time t, const Array &x) const
 returns the diffusion part of the equation, i.e. $ \sigma(t, \mathrm{x}_t) $
Disposable< Arrayapply (const Array &x0, const Array &dx) const
Disposable< Arrayevolve (Time t0, const Array &x0, Time dt, const Array &dw) const
Real v0 () const
Real lambda () const
Real omega () const
Real alpha () const
Real beta () const
Real gamma () const
Real daysPerYear () const
const Handle< Quote > & s0 () const
const Handle
< YieldTermStructure > & 
dividendYield () const
const Handle
< YieldTermStructure > & 
riskFreeRate () const
Time time (const Date &) const


Detailed Description

Stochastic-volatility GJR-GARCH(1,1) process.

This class describes the stochastic volatility process governed by

\[ \begin{array}{rcl} dS(t, S) &=& \mu S dt + \sqrt{v} S dW_1 \\ dv(t, S) &=& (\omega + (\beta + \alpha * q_{2} + \gamma * q_{3} - 1) v) dt + (\alpha \sigma_{12} + \gamma \sigma_{13}) v dW_1 + \sqrt{\alpha^{2} (\sigma^{2}_{2} - \sigma^{2}_{12}) + \gamma^{2} (\sigma^{2}_{3} - \sigma^{2}_{13}) + 2 \alpha \gamma (\sigma_{23} - \sigma_{12} \sigma_{13})} v dW_2 \ \ N = normalCDF(\lambda) \\ n &=& \exp{-\lambda^{2}/2} / \sqrt{2 \pi} \\ q_{2} &=& 1 + \lambda^{2} \\ q_{3} &=& \lambda n + N + \lambda^2 N \\ \sigma^{2}_{2} = 2 + 4 \lambda^{4} \\ \sigma^{2}_{3} = \lambda^{3} n + 5 \lambda n + 3N + \lambda^{4} N + 6 \lambda^{2} N -\\lambda^{2} n^{2} - N^{2} - \lambda^{4} N^{2} - 2 \lambda n N - 2 \lambda^{3} nN - 2 \lambda^{2} N^{2} \ \ \sigma_{12} = -2 \lambda \\ \sigma_{13} = -2 n - 2 \lambda N \\ \sigma_{23} = 2N + \sigma_{12} \sigma_{13} \\ \end{array} \]


Member Function Documentation

Disposable<Array> apply ( const Array x0,
const Array dx 
) const [virtual]

applies a change to the asset value. By default, it returns $ \mathrm{x} + \Delta \mathrm{x} $.

Reimplemented from StochasticProcess.

Disposable<Array> evolve ( Time  t0,
const Array x0,
Time  dt,
const Array dw 
) const [virtual]

returns the asset value after a time interval $ \Delta t $ according to the given discretization. By default, it returns

\[ E(\mathrm{x}_0,t_0,\Delta t) + S(\mathrm{x}_0,t_0,\Delta t) \cdot \Delta \mathrm{w} \]

where $ E $ is the expectation and $ S $ the standard deviation.

Reimplemented from StochasticProcess.

Time time ( const Date  )  const [virtual]

returns the time value corresponding to the given date in the reference system of the stochastic process.

Note:
As a number of processes might not need this functionality, a default implementation is given which raises an exception.

Reimplemented from StochasticProcess.