This repository provides a GPU-accelerated implementation of Monte Carlo methods for pricing European options under the Heston stochastic volatility model, developed as part of the Master's course "Massive Parallel Programming on GPU Devices for Big Data" from the Probabilités et Finance Master's program at Sorbonne Université.

The project includes implementations of three discretisation schemes:

• Euler scheme
• Exact scheme (Broadie-Kaya)
• Almost Exact scheme (Haastrecht-Pelsser)

Each scheme leverages CUDA for high-performance parallel computations. Trajectories are simulated independently at thread-level, with efficient shared-memory reductions performed at the block level.

heston.cu: CUDA source code for the simulation.

heston_gpu.pdf: Project presentation slides.

This implementation was inspired by the CUDA examples and materials provided by Lokman Abbas-Turki as part of the course "Massive Parallel Programming on GPU Devices for Big Data".