Hedge Fund Replication with Deep Neural Networks and Generative Adversarial Networks
Abstract
Hedge fund replication is a means for allowing investors to achieve hedge fund-like returns, which are usually only available to institutions. More traditional money managers would like to offer hedge fund-like returns to retail investors by replicating their performance. There are two primary challenges with existing hedge fund replication methods, difficulty capturing the nonlinear and dynamic exposures of hedge funds with respect to the factors, and difficulty in identifying the right factors that reflect those exposures. My research investigates Deep Neural Networks and Generative Adversarial Networks to address the challenges of factor-based hedge fund replication and improves on previous results.
Fields of Study & Subjects
Stock Market Forecasting Methods
Computational Physics and Python Applications
Complex Systems and Time Series Analysis
Publication details
- Journal / source
- Journal of Advanced Robotics Autonomous Systems and Human-Machine Interaction
- Publisher
- International Prime Publications Inc
- Volume
- 02
- Issue
- 03
- Pages
- 01–15
- Publication date
- 10 August 2026
- Language
- EN