Hedge Fund Portfolio Construction Using PolyModel Theory and iTransformer

TL;DR

Facing sparse hedge fund time series, this work combines PolyModel theory with an iTransformer-based forecasting framework to construct hedge fund portfolios. PolyModel builds a large, nonlinear feature space from a wide risk-factor pool using Hermite-polynomial regressions and Ridge estimation, producing metrics such as $LTA$, $LTR$, and $SVaR$ that feed learning. The iTransformer then predicts trend probabilities for each fund, enabling simple average (SA) and weighted average (WA) portfolios, which are rebalanced monthly. Empirical results on data from 1994-2023 show that SA/WA outperform two HFR benchmarks in terms of cumulative returns and risk-adjusted measures, demonstrating the practical value of integrating domain knowledge with deep learning for high-dimensional time series in finance.

Abstract

When constructing portfolios, a key problem is that a lot of financial time series data are sparse, making it challenging to apply machine learning methods. Polymodel theory can solve this issue and demonstrate superiority in portfolio construction from various aspects. To implement the PolyModel theory for constructing a hedge fund portfolio, we begin by identifying an asset pool, utilizing over 10,000 hedge funds for the past 29 years' data. PolyModel theory also involves choosing a wide-ranging set of risk factors, which includes various financial indices, currencies, and commodity prices. This comprehensive selection mirrors the complexities of the real-world environment. Leveraging on the PolyModel theory, we create quantitative measures such as Long-term Alpha, Long-term Ratio, and SVaR. We also use more classical measures like the Sharpe ratio or Morningstar's MRAR. To enhance the performance of the constructed portfolio, we also employ the latest deep learning techniques (iTransformer) to capture the upward trend, while efficiently controlling the downside, using all the features. The iTransformer model is specifically designed to address the challenges in high-dimensional time series forecasting and could largely improve our strategies. More precisely, our strategies achieve better Sharpe ratio and annualized return. The above process enables us to create multiple portfolio strategies aiming for high returns and low risks when compared to various benchmarks.

Figures (1)

• Figure 1: This figure plots the cumulative returns of the 4 strategies.