Jurnal PREDIKSI HARGA OPSI DENGAN KOMPUTASI PARALEL SOLUSI NUMERIK PERSAMAAN HESTON MENGGUNAKAN METODE FINITE DIFFERENCE PADA GRAPHICS PROCESSING UNITS

No view

Jurnal PREDIKSI HARGA OPSI DENGAN KOMPUTASI PARALEL SOLUSI NUMERIK PERSAMAAN HESTON MENGGUNAKAN METODE FINITE DIFFERENCE PADA GRAPHICS PROCESSING UNITS

Wikanargo, Matheus Alvian

(2019)

OPTION PRICING PREDICTION WITH COMPUTATION OF PARALLEL NUMERICAL SOLUTIONS OF HESTON EQUATION USING FINITE DIFFERENCE METHOD IN UNITS GRAPHICS PROCESSING.

S2 thesis, UAJY.

Abstract

Options are financial instruments in which two parties agree to
exchange assets at price or strike and the date or maturity has been determined
previous. Options provide investors with information to set strategies so
can increase profits and reduce risk. Option prices can be valued
using the popular Heston equation model used. Equation model
Heston has advantages compared to other equation models because
assuming volatility is not constant with time or stochastic volatility. Volatility
which is not constant with time according to reality because of the underlying asset
as the base can experience fluctuations. This equation has a weakness
because it is a derivative equation. The derivative equation is an equation
which is difficult to solve, one way to solve derivative equations
easily is using numerical solutions. Numerical solutions can
solving derivative equations but requires a heavy computational process
and slow.
The numerical solution to the finite difference method of non-uniform grids is
methods can be used flexibly and do not require processing
matrix. Heston equations can be solved by nonuniform finite difference methods
grids because the Heston equation can be assumed as an equation
parabolic. Numerical solutions require heavy computational processing time
and slow, because there are many elements of calculation and iteration. Processing
numerical computing solutions can be done using parallel programming
Compute Unified Device Architecture (CUDA) to speed up the process.
Valuation of option prices needs to be processed accurately according to reality and fast,
so that the resulting value can be utilized at the best momentum.
This study proposes a numerical solution to the non-uniform finite difference method
grids to solve the Heston equation model to get the results
accurate and fast.