Approximating Optimal Asset Allocations using Simulated Bifurcation

TL;DR

This paper applies Simulated Bifurcation algorithms to efficiently approximate optimal asset allocations for large asset pools, demonstrating a Python implementation on S&P 500 assets and addressing sub-allocation selection.

Contribution

It introduces a novel application of Simulated Bifurcation to finance, providing an efficient method for large-scale asset allocation problems with a practical Python implementation.

Findings

Effective approximation of optimal asset allocations for 441 assets

Fast solution times for sub-allocation selection

Demonstration of physical principles in financial optimization

Abstract

This paper investigates the application of Simulated Bifurcation algorithms to approximate optimal asset allocations. It will provide the reader with an explanation of the physical principles underlying the method and a Python implementation of the latter applied to 441 assets belonging to the S&P500 index. In addition, the paper tackles the problem of the selection of an optimal sub-allocation; in this particular case, we find an adequate solution in an unrivaled timescale.