PARTNER CONTENT
By Robert D Stock, PhD
Principal, Analytics Research and Equity Research, SimCorp
When I used to run due diligence on hedge funds, I met a trader from a hedge fund running a long-short market-neutral strategy. The fund was neutral to various risk factors and had to generate ‘pure alpha’. The trader I met was doing really well while the other PMs were losing money. How? It was around the dot-com bubble and the hedge fund trader noticed that a group of seemingly unrelated companies were moving together – both up and down – and neutralized his portfolio to this new “internet” factor.
Can you do this on a more systematic basis?
Statistical and fundamental factor risk models can help hedge funds – from statistical arbitrage to market neutral – understand risk factors and exposures. Statistical models identify unexpected sources of risk not captured by fundamental models. While powerful, they lack intuitive explanatory power, whereas linear fundamental models provide an understandable story but are limited by their factor set, potentially resulting in more unexplained residual risk. The standard assumption is that this fundamental residual risk is random and unstructured. Or is it?
By taking these residual security returns from our fundamental model, the Axioma US Equity Factor Risk Model (US5.1) finds a neural network-like polynomial structure based on our model’s factors using Instrumented Principal Components Analysis, a machine learning technique. This delivers neural network-like power to uncover hidden risks but through an intuitive factor lens, instead of an opaque black-box. This new factor is something we call the Non-linear Residual Structure.
How can hedge funds spot non-linear residual factors?
The residual structure of non-linear terms is driven by investor preferences and will change over time. In the chart below, you can see how some of the common important terms (Size Squared, Size Cubed and Size x Residual Volatility) have done just that.
By monitoring the environment, the Non-linear Residual Structure factor can find hidden risks and connections among other factors, which can be beneficial to hedge funds. For example, we found that until about 2002, positive exposure to the Non-linear Residual Structure factor would have been the right call, but then the opposite would have been true, especially since 2021. And, for those willing to make macro directional bets, this factor could also provide an additional source of return.
Figure 1: Sample of common Non-linear Residual Structure factor terms
Source: SimCorp, Axioma US Equity Factor Risk Model 5.1
Introducing a new era for sentiment analysis
This new factor brings together the explanatory power of fundamental models with the dynamic combinatorial power of statistical models. It is part of a family of market-based sentiment factors in the latest version of our US risk model and includes other hedge-fund relevant factors including Hedge Fund Crowding, Short Interest and Opinion Divergence.
Learn more about the latest Axioma US Equity Factor Risk Model at www.qontigo.com/US5-1.
For more about our hedge fund-related solutions, go to www.qontigo.com/hedge-funds
Robert (Bob) D Stock, PhD, Principal in Equity and Analytics Research, SimCorp – Bob develops Axioma factor risk models for SimCorp. After earning his undergraduate degree in Physics from Princeton and his doctorate in Physics from Carnegie Mellon, Bob worked for nearly a decade at MIT’s Lincoln Laboratory in the Directed Energy Group, performing simulations and analysis of high energy laser beam propagation through the atmosphere for missile defense applications. Bob then served the Ultra-High Net Worth segment for 15 years, first as a Quantitative Researcher for a boutique outsourced-CIO where he developed the majority of the firm’s intellectual property for hedge fund analysis, risk management, forecasting, and asset allocation. Subsequently, he served as a systematic portfolio manager for a large Trust before joining Axioma (now part of SimCorp) in 2022.
