Skip to content
Contact us

Why the risk factor model is the hedge fund’s unsung hero

4 use cases for factor-based models

It’s no secret that hedge funds are one of those most secretive players in our industry.  

Authors:

Duncan Coutts,
Axioma Product Specialist 
  
Saboor Zahir,
Axioma Product Specialist 

After all, if they don’t protect their ideas and intellectual property, they could lose their competitive advantage – not to mention the trust of their investors. One thing that isn’t a secret is that hedge funds rely on a heavy tech stack, from front to back. And while execution algos and AI might be the first thing to come to mind, it’s the risk models (or rather, the Axioma factor-based risk models that are the top picks for our hedge fund product specialists. Why? We asked two members of the team, Duncan Coutts and Saboor Zahir to lift the lid on how our hedge fund clients are using risk factor models to solve some of the trickiest challenges.

1: Factor-based risk models for accurate decision-support that reflects a non-traditional perspective on the market    

[Duncan] Often, I speak with hedge fund managers who extract value by virtue of perceiving the market landscape in specific ways. It’s generally accepted that each country or region has some specificity – that they behave differently to each other – but this principle extends to any definition of the ‘market’ one cares to choose. For example, sectors and themes are equally valid ways to assign commonality for a group of assets. From these alternative perspectives, the definition of ‘average exposure’ (and the dispersion) for characteristics such as size, momentum or value, and their returns and volatilities, can be markedly different from the baseline given by a standard risk factor model.

This means for a new risk number and decomposition – which are statistically more accurate for portfolios investing in the theme of interest – the specific investor can use the gap analysis or ‘risk spread’ versus a standard risk factor model to identify opportunities, corroborate decision-making, and enable fairer comparisons of portfolios and assets from distinct categories. Importantly for managers, global coverage can be retained, and it is just the focus of estimation that changes.

By using the Axioma Risk Model Machine in combination with the Axioma range of equity factor-based risk models, hedge funds can build their own unique risk models to complement their off-the shelf ones. By covering alternate perspectives, this can prevent otherwise hidden risk-taking, help refine the research process and ensure a more specific allocation of risk to the premia that the hedge fund manager believes will outperform.

2: Factor-based risk models for advising our portfolio managers more proactively and to increase collaboration

[Saboor] Traditionally, risk management has been reactive, flagging risk limit breaches and non-compliant trades. But at hedge funds (and increasingly within asset management), quantitative risk managers have a collaborative relationship with portfolio managers, providing them with detailed guidance that adds support to the good decisions as well as preventing problems.  

This form of proactive risk management is where the Axioma risk factor models come into play. Regardless of asset class type – from equities, fixed income or somewhere in between – the Axioma models enable a dialogue between risk managers and portfolio managers in which the common factors of the model act as a framework for monitoring, understanding, and discussing how the market is (or could) influence the portfolio. Portfolio managers like it because the conclusions of a factor analysis are immediate and actionable and can be interpreted and sliced at all levels of a portfolio from single name assets, upward. The risk managers like it because the factors facilitate more than risk measurement; it is also performance attribution, scenario generation and example hedge construction, all of which complement their guidance. Having a ‘nowcast’ of the portfolio in full context of the market it exists within is a win-win for all concerned. Decision making becomes more agile, and with a robust and proactive oversight, the portfolio manager can focus more energy on strategic research.  

3: Factor-based models for a new perspective on the risk and return of fixed income strategies

[Duncan] Fixed income data has always been notoriously difficult to work with and has stumped many who have tried building curves and fixed income risk models for their credit strategies, in-house. Traditionally, vendors relied on ‘granular models’ where the issuer spread returns alone drive a vast covariance matrix. The issue with this is that it is unwieldy and assumes a lot about the accuracy of individual issuer correlations. In contrast, the equity world has long since imposed systematic structure (i.e., factors) onto asset returns. This helps to identify statistically testable commonalities (of countries, industries, styles) and isolate idiosyncratic risk and return and at the same time reduce risk to a manageable dimension and gain more flexibility over the model’s reactivity. 

Some hedge funds saw the step-change potential but knew the quality of curves required to support such a model, together with the style factor development would be years of painstaking effort – work that the research team for Axioma solutions had already undertaken. We have been able to help existing and new hedge fund clients when their existing fixed income models lacked the detail or flexibility to decompose credit risk into significant factors like market, sectors, quality, and fundamentally (no pun intended!) revitalize their perspective on how the market is influencing their strategies. 

The Axioma Credit Spread Factor Risk Models are derived from a cross -sectional regression of issuer spread returns with Duration Times Spread-based factor exposures. This proprietary methodology gives portfolio managers and risk managers an additional lens to better understand credit risk and can be viewed through a parsimonious factor approach or a granular issuer-based approach.  

4: Factor-based factor risk models for hedging and neutralizing risk within a multi-asset portfolio

[Saboor] With the introduction of fewer, but more intuitive and explanatory factors to explain the total market, targeted exposure and risk hedging becomes feasible. The combination of a suitable model and optimization strategy enable this kind of decision process.  

Dealing with cross-asset risks is a great example and we often speak with managers seeking to remove risks that confound their investment theses. For example, if one wants equity exposure to quality banking and energy earnings but without the underlying interest rate and oil price volatility impacts on these sectors, it would be natural to ask how many (and which) bond and oil futures are needed to negate the broader macroeconomic environment implicit in the equity book. Managers do not want to take directional bets on oil and interest rates because that may not be their area of expertise.   

The cross-asset risk factor model answers the question of how correlated interest rates and oil to the factors that drive the risk in these companies. By using a multi-asset class risk model in conjunction with Axioma Portfolio Optimizer, the manager can effectively reduce the risk of their portfolios to oil and interest rates.

The Axioma Global Multi-Asset Class (MAC) Risk Model incorporates a consistent and parsimonious factor structure to align to a hedge fund’s overall portfolio. With broad market coverage across equities, fixed income, and commodities, the MAC factor-based risk model gives a true holistic picture of the overall portfolio.  

Serving a wide range of hedge fund strategies  

From decision-support to bringing portfolio managers and risk managers closer together, factor risk models have played a significant role for our hedge fund clients. Axioma multi-asset solutions cover a wide range of hedge funds strategies, including but not limited to:  

  • Equity Long-Short 
  • Global Macro 
  • Credit Long-Short  
  • Relative Value 
  • High Yield  
  • Market Neutral 
  • Quantitative 
  • Fund of Hedge Funds  

In addition to equity, fixed income and multi-asset factor risk models, Axioma solutions can help solve the challenges of hedge funds including strategy building and optimization and uncovering arbitrage opportunities. Axioma tools are just some of the solutions available on SimCorp One, the platform that powers decisions at the speed of now.

View our hedge fund solutions.

  • Privacy policy
  • Cookie Policy
  • Terms of Use
  • Trademark guidelines

Copyright © 2024 SimCorp A/S