Calculated Risks
Collaborative Intelligence
Research

Research

Our approach

We believe that today’s research is tomorrow’s performance. Research is an integral part of portfolio managers’ work and we have a long tradition of applying scientific research methods to create innovative strategies for our clients. Our collaboration-based approach to research has led us to build a strong, extended network of industry practitioners and academics to generate and develop ideas.

Visit our dedicated Insights section to find out more about our research.

Today’s research drives tomorrow’s performance

1995

Minimum Variance research project, on Prof. R. A.Haugen research

2001

Behavioural finance risk questionnaire

2004

First integration of ESG factors

2007

Incorporation of extreme and asymmetrical risks in optimization

2011

Development of a proprietary statistical multi-factor risk model

2013

'Private Assets Allocator', an tool driven by quantitative an non-quantitative expectations

2014

Launch of a multi-factor equity strategy

2014

Development of our Macro Risk-Based investment process (FRAME) and associated tools such as Nowcasters

2015

“PE-Risk Premia”, liquid proxy exposure to a portfolio of private companies risk

2015

Roll-out of range of Alternative Risk Premia Strategies

2016

Proprietary CO2/climate risk analysis methodology, relying on TruCost Data

2019

AI-powered scoring tools for private equity fund and company investments

2019

Machine Learning model, allowing better detection of stocks with increased downside risk

2020

Long Term Vega added as part of systematic defensive strategies

Delivering research Into strategies

Project

Objective

Outcome

Evolving Nowcasters

Integrating News through Natural Language Processing methodologies into our macro-signals indicators

Seek to improve the reactiveness of our Nowcaster macro indicators by adding macro news sentiment signals

Implementation of the ‘Newscaster’ sentiment signals alongside the ‘Nowcaster’ signals in our multi-asset investment processes

Machine Learning in Private Equity

Applying machine learning algorithms to forecast the return of Private Equity funds based on structured data, unstructured date and GP (General Partners) sentiment

Score all known funds in fundraising

Efficient shortlisting of funds to be reviewed by analysts:

  • 700 funds trained
  • 300 funds tested
  • 70% accuracy achieved

ESG Scoring

Designing a proprietary ESG scoring methodology for companies

Develop an ESG indicator aligned with our equity investment process around active risk management

Implementation of monthly indicators in our equity investment processes from the risk perspective. We took into account differences in materiality of ESG issues for different sectors and incorporated the impact of controversies, when relevant.