Historically, over the long term, economic growth has provided an average annual return of around 7% from a balanced portfolio spread across markets. Using machine learning algorithms we aim to achieve in excess of this, after costs, with reduced volatility.
We continually enhance machine learning to optimise investments. Signal generation, asset allocation and risk management evolve through artificial intelligence applied to expanding data sources. This improves stock selection, portfolio construction and downside protection.
The Financial Science automated trading engines rapidly process very large volumes of data, optimise investment strategies and dynamically account for alterations in the market. They use an ensemble approach across a range of data sources. By processing more information more effectively we provide investors with a lower risk, higher return edge.
Our approach to investments originated from University College London’s Doctoral Training Centre in Financial Computing and Analytics through an initiative to commercialise research and demonstrate best practice in the field.
OPERATIONALLY WE TARGET
LOW FEES
Through low costs and low margins, aligned with investor returns
EFFICIENT CODE
Developed on a robust scalable architecture
MODULAR SYSTEM
Allows rapid deployment of new insights across multiple markets
COMPREHENSIVE STATISTICAL TECHNIQUES
To test, evaluate and monitor signals and strategies
TRANSPARENT ARCHITECTURE
Provides an intuitive view of investment process, performance drivers and risk exposures
CONTINUOUS R&D
To maintain an IP advantage in the markets
RISK MANAGEMENT
Captures inefficiencies and works to mitigate catastrophic hazards
We present our key holdings and research findings to our investor partners transparently as part of our commitment to ongoing learning and improvement.
Our Team and Advisory Board includes leading experts from academia and industry. We support a much more open industry for investors.