Former founder of Ikos, Dr Martin Coward, considered by many within the hedge fund industry to be one of the true pioneers of systematic electronic trading, is opening his new quant fund to investors.
Based in Malta, dormouse is regulated by the Malta Financial Services Authority (MFSA and is a registered Commodity Trading Advisor (CTA) and Commodity Pool Operator (CPO) under the US National Futures Authority.
Over the past 2.5 years the team has honed the trading strategy and the dormouse technology platform, steadily building assets, which currently stand at USD270m. Now the fund is ready to open up its doors to investors with Dr Coward noting that the strategy has scale of USD1bn but that the firm is in “no hurry” to reach that target.
In an official press release, Dr Coward comments that the fundamental objective at dormouse is to “exploit the increased correlation [between asset classes] whilst simultaneously neutralising it to generate uncorrelated alpha”.
Correlations over the last decade have strengthened, Dr Coward tells Hedgeweek, as large market participants (hedge funds, SWFs) increasingly trade similar assets and strategies.
“Higher correlations result in increased systemic risk, but also in reversionary and cross-asset phenomena. Some of these phenomena can be isolated by modelling spread relationships, namely the spread between a security (or securities) and these systemic risk factors. Because most funds attempt to generate alpha by timing systemic risk factors, the alpha we generate is uncorrelated,” says Dr Coward.
The investment target returns for investors are 15 per cent annualised. Strategy-wise, the program developed by dormouse, which will not be reliant on trend following, will focus on identifying under or overpriced liquid securities across stock indices, fixed income, rates, FX and commodities. It can best be categorised as a systematic multi-strategy fund.
Identifying unconventional sources of alpha will be the end game, with Dr Coward stating that making dormouse different from other funds “is our primary objective. Our risk management systems are fully automated. The risk ratio fluctuates around the average annualised risk level with risk scaling using an adaptive covariance matrix. Less exposure is made to ‘risk premium’.”
Risk will be allocated to any strategy with a sufficient risk-reward ratio that can be supported by scientific analysis and can provide investors with diversified returns.
“The scientific approach is really the core of dormouse's philosophy,” explains Dr Coward. “We have rejected strategies with attractive returns that don't have a conceptual basis that can be supported by data and scientific analysis. We have even removed strategies from the portfolio that initially seemed to be supported by the data but later were found to be lacking when further data became available or further analysis raised questions that couldn't be answered scientifically.”
Currently, the program trades 40 of the most liquid futures markets although ETF and equities strategies are also being investigated. Dr Coward expects some of these strategies to pass muster scientifically and from a correlation and risk-reward standpoint “and will make it into the live portfolio in 2014”.
With respect to how dormouse approaches risk management, each strategy employs its own internal risk management controls to ensure it contributes well-regulated risk to the overall portfolio. The portfolio then allocates a roughly equal share to each of the five asset classes.
“Individual positions are capped based both on risk and on notional amount. Since our execution is fully automated, we also manage operational risk with multiple techniques. More broadly, an element of our scientific approach requires simplicity of implementation – this helps us to achieve robust strategies and execution algorithms.
“Ultimately, the most effective risk management is having the monitoring systems in place to keep a close watch on all aspects of the system's functioning and having eyes and brains watching it,” says Dr Coward.
