Algorithmic complexity in the heston model

Abstract

In this paper, we present an in-depth investigation of the algorithmic parameter influence for barrier option pricing with the Heston model. For that purpose we focus on singleand multi-level Monte Carlo simulation methods. We investigate the impact of algorithmic variations on simulation time and energy consumption, giving detailed measurement results for a state-of-the-art 8-core CPU server and a Nvidia Tesla C2050 GPU. We particularly show that a naive algorithm on a powerful GPU can even increase the energy consumption and computation time, compared to a better algorithm running on a standard CPU. Furthermore we give preliminary results of a dedicated FPGA implementation and comment on the speedup and energy saving potential of this architecture.

Towards Automated Benchmarking and Evaluation of Heterogeneous Systems in Finance

“…We have employed the proposed benchmark to evaluate the performance of different algorithmic tunings in a Multilevel Monte Carlo (MLMC) simulations [20]. Furthermore, have used the benchmark to validate and characterize our FPGA architectures for European barrier option pricing in the Heston model [6,9].…”

“…In addition, we need to consider algorithmic flavors that may be available for a particular market model. For example, Marxen et al have shown that log price simulation with full truncation and continuity correction performs well for European barrier options in the Heston model, but this also depends on the specific market and option parameters [20]. Figure 4.1 illustrates the large design space for constructing an option pricing system.…”