CFM started as a hedge fund manager focusing on systematic trading strategies. Today such strategies are referred to as absolute return of Alpha strategies.
Our two flagship investment programs were launched in 1991 and 2003 respectively. They are multi award winning and have the objective of achieving consistent returns, in variable market conditions with a risk profile that seeks to be less volatile than general market indices. The performance of these programs, which aims to have less correlation to general market indices, represents our Alpha strategies. These strategies are based on exploiting medium term (days to months) inefficiencies in markets and so portfolio turnover is relatively high.
These programs are the zenith of our collective capability and are structured to reflect the fact that achieving a combination of acceptable risk and decorrelation from market indices requires considerable expertise, investment and R&D capability. The strategies in the programs are therefore subject to constant evolution with a view to maintaining the Alpha characteristics and manage decay.
For over 25 years we have invested in quantitative research and development resulting in the creation of successful absolute return programs. Alpha strategies are by definition capacity constrained, complex and have associated higher fees making such strategies attractive to only a small minority of investors. Mainstream investor appetite for simpler strategies with an attractive fee structure is growing. Mindful of this and the constraints of the absolute return programs, in 2012 we began developing a different type of program, which would use the firm’s existing expertise and capital expenditure to appeal to a broader set of investors. The programs were launched in 2013 and 2014 with the hallmarks of our quantitative approach but with less complexity, higher correlation and appropriate fees.
Alternative Beta, a definition: ‘Alternative strategies’ have historically been associated with hedge funds on account of the fact that the traded instruments in such strategies are often not ‘traditional’ equities and bonds and the returns profile is often decorrelated from equity and bond benchmarks. Such strategies are now making their way into more mainstream investment circles, but the associated investment styles continue to be known as ’alternative’. Alternative Beta is therefore an investment style with returns correlated to these now well-known alternative benchmarks and therefore not to traditional equity or bond benchmarks such as the S&P 500. The definition of an alternative benchmark is not as clear cut, or accessible, as in the case of equity or bond indices, but is instead assumed to be a standard implementation of a standard alternative trading strategy.
While a strict definition of Alternative Beta strategies remains elusive, various important themes tend to emerge:
- Strategies that persist over long periods of time – often decades or even centuries. Levels of statistical significance need to be sufficiently high to establish strong conviction that the systems are not over-fitted but well anchored in the structure of the markets.
- Strategies that are slower moving, as opposed to the short term inefficiencies that are generally arbitraged away through time and exploited in an absolute return strategy. Importantly the slower strategies tend to be scalable to higher capacity.
- Strategies that are explainable, understandable and plausible.
- Alternative beta strategies can generally be split into two types – risk rewarding or so called risk premia and pure market anomalies often of behavioural origin.
In short, a diversified Alternative Beta program is a mix of simple, well-justified strategies, seeking to deliver persistent excess returns with scalable capacity, while exhibiting low correlation to traditional equity and fixed income benchmarks.
The synergies between the Alternative Beta and our Alpha programs make for a highly differentiated approach when it comes to implementing the Alternative Beta strategies and one that shouldn’t be underestimated. This differentiation manifests itself in the following ways:
Portfolio construction, together with return generation and execution, is a vital aspect of building a sound investment program. It requires robust data and developed statistical methods which we have honed over many years.
Systematic risk management based on in-house IT infrastructure
Our IT/Data infrastructure, built over two decades at a cost of more than €100m, has been specifically designed for quasi real time risk evaluation. So while the trading patterns are slower in the Alternative Beta strategy, regular updates of positions and market condition indicators are important in building a reliable risk control infrastructure.
Robust operational risk control
The monitoring and management of operational risk at all levels in the production chain is important. While certain risk monitoring tools are embedded in the trading systems and maintained by the front office teams, we also maintain a dedicated, independent risk team, reporting directly to our firm’s directors, whose mandate is to independently validate financial risk estimates and to independently impose operational risk limits.
Industrialised, large-scale data processing
Our data team is responsible for collecting, cleaning, manipulating, and managing terabytes of incoming data every day. Our data sets range from price-related information (e.g., prices, implied volatilities) to fundamentals (e.g., corporate financial statements) and non-financial information for trading ideas that exploit idiosyncratic, market-specific inefficiencies. Given the lower turnover of Alternative Beta strategies, achieving statistical significance is facilitated by the longest back tests possible. Time-series data for futures and equities extends back at least to the 1960s and 1970s and, where possible, we also maintain certain data series as far back as 1800 (e.g., monthly data for many indices, commodities, bonds, and various interest rates). We currently monitor approximately one million instruments.
The Importance of minimising costs
Controlling trading costs in the implementation of any strategy is critical. Our dedicated execution research team manages our infrastructure and seeks to model, measure and reduce costs. We have an extensive database of the execution of our own trades, data which is not commercially available, and which provides valuable insight into the dynamics of price impact. Our execution team has also published numerous papers in this field, in particular on the subject of market impact.
We believe that financial markets are constantly evolving and that the markets have a degree of predictability made increasingly more apparent by data. This creates a strong environment to create alternative investment strategies that provide risk adjusted returns with a low correlation to traditional investment markets benchmarks.
We believe that:
- Most alpha trading models decay over time
- Delivering continued value is only maintained by an ongoing research and development effort
- Risk management is of utmost importance
- Diversification works
Over half of our employees are data scientists and since our inception in 1991, we have invested over €100m in sophisticated information technology systems. This state of the art platform gives vast processing power to collect and sort the data needed by the research teams. Today, we have over 1,500 servers enabling the collection and presentation of over 3 terabytes of information every day, the equivalent of a typical academic research library.
The data team collects, cleans, and manages terabytes of incoming data every day. We thrive on new forms of data as inputs to systematic trading ideas. Data is also collected to simulate the whole production chain from end to end to ensure that positions and performance are in line with back tests. Data sets need to be as broad as possible to inspire new ideas, but also reliable and readily accessible to the research team.
Systematic trading also requires advanced technological capability whether this is placing orders, reconciliation or monitoring risk. We are connected in real-time to about 20 counterparties (exchanges, execution brokers, electronic platforms, banks, etc.). We have developed control systems to monitor operations and manage potential issues.
As a firm, research is in our DNA and the basic principles established in 1991 remain unchanged today. We create models based on the statistical properties of financial instruments and then use these models to develop systematic trading strategies.
We have assembled a team of top researchers with strong academic track records. Most of our researchers hold PhDs, from some of the world’s leading academic institutions, mostly in physics or other hard sciences. We carefully select our research team ensuring that most of them join us directly from academia, a distinct point of difference for the firm. This gives us a uniquely dispassionate and unbiased approach to our research, which is ultimately reflected in the strategies across our programs. In particular this is important in our Alpha programs, where the researchers’ distinctive backgrounds are critical in order to drive the consistent innovation and R&D necessary for maintaining the programs’ alpha characteristics.
All trading strategies are rigorously tested and piloted prior to deployment. A dedicated team monitors trading portfolio risk to gauge risk levels and performance. This process relies on in-house developed monitoring software. In periods of extreme market volatility or unprecedented market events, the board has the authority to override the algorithms and reduce portfolio risk. We manage other risks and in addition, we have built over more than two decades professional functions including: IT, HR, legal and compliance to support the firm and its clients.