100% rule based investment protocol with fully automated decision making process. This systematic approach ensures full transparency in the selection of securities and construction of diversified portfolios. In addition, the rule-based process can ensure repeatable results and thus more consistent alpha.
1. Methodical monitoring of global stock universe
Starting from a universe of 15.000 stocks globally, SIMAG’s proprietary systems screens around 7.500 liquid stocks for exploitable patterns. For our mutual funds, we focus on ESG-compliant investment universes, such as defined by the range of MSCI ESG Leaders indices following a Best-in-class approach – the indexes select the highest rated companies in each sector and exclude companies involved in severe controversies.
2. Scoring of stocks based on proprietary machine learning algorithms
Daily, the proprietary artificial intelligence engine applies 90 indicators in 5 models with different time horizons, each using 500 decision trees. Within the 5 models, a SIMAG proprietary AI algorithm connects the indicators via "if-then" rules, depending on the thresholds the AI algorithm has learned from historical data. All scores across models and trees are averaged to make the final score (SIMAG score) more robust and less noisy.
3. Stock selection and optimal portfolio construction
Every day, stocks are ranked according to their scores (from highest to lowest) which forms the basis of the models over / underweight decisions. The larger the score the more attractive a stock is for inclusion in the portfolio according to the model. The result is a diversified portfolio across sectors (and countries).
4. Risk management & system integrity
We have strong governance and independent risk oversight processes in place with daily, weekly and monthly risk monitoring. Our Research & Development Committee oversees and approves al changes in the investment process before deploying to the portfolios.