Intelligent Lattice Search: Efficiency in Option Pricing using Python, C++, Cython and Numba

[@ranjithramachandran8009]

Very informative. I was wondering if it is possible to further enhance the execution time for the baseline code using JAX. Can that be implemented ? I do understand that it is ideal for large computations (GPU) and may not perform well (can become slow) in comparison to Numpy. JAX with lax.scan for iterative computations like option pricing. However it is questionable if JAX can beat the Numba execution time.

[@BrianByrneFinance]

Hi @ranjithramachandran8009, Thank you for the great suggestion! JAX with lax.scan could be an interesting approach for iterative computations like option pricing, especially with GPU acceleration. As you noted, JAX may not always outperform Numba on CPU-bound tasks, but it’s worth testing. I’ll consider exploring this comparison in a future video-thanks again for your thoughtful input! best, Brian

[@laurentbouchard1448]

Numba can be set up to use GPUs; you get the speedup benefits if you have to value lots of contracts is my understanding. I'v been running this code with the Intel SVML/fastmath=true Numba parameters and the speed is ridiculously fast. It really is amazing.

[@BrianByrneFinance]

@@laurentbouchard1448 Dear Laurent, Do you have a colab link to that? If so could you share? best, Brian