Machine Learning (Deep) Application to Portfolio Allocation

This study investigates the application of machine learning and deep learning techniques to portfolio allocation, with particular emphasis on their ability to capture complex and dynamic relationships among financial assets. The study compares recurrent neural network architectures, including Recurrent Neural Networks (RNNs), Long Short-Term Memory (LSTM), and Gated Recurrent Units (GRU), together with Transformer models, against classical portfolio optimization approaches represented by the Markowitz mean-variance model and the Black-Litterman model. The empirical analysis uses daily financial data for ten highly liquid assets: Apple (AAPL), Microsoft (MSFT), Alphabet (GOOGL), Amazon (AMZN), Tesla (TSLA), NVIDIA (NVDA), Meta Platforms (META), Netflix (NFLX), JPMorgan Chase (JPM), and Bank of America (BAC). The analysis covers approximately eight years of historical data and incorporates selected macroeconomic indicators obtained from the Federal Reserve Economic Data (FRED). Portfolio performance is evaluated using measures including annualized return, volatility, Sharpe ratio, and maximum drawdown. The study finds that the performance of deep learning approaches varies substantially across architectures. In the reported experiments, Transformer and RNN-based approaches demonstrate strong portfolio performance relative to the classical benchmarks, while the GRU and LSTM models do not consistently outperform the traditional approaches. The study contributes to the growing intersection of artificial intelligence, machine learning, and computational finance by examining how deep learning architectures can be incorporated into portfolio allocation. It also highlights the importance of model selection, robust evaluation, and careful interpretation when applying machine learning methods to financial markets. This work was completed as an MScFE Capstone Project at WorldQuant University and is intended for academic and research purposes. It does not constitute investment or financial advice.