Minimum Variance Solutions

A passion to invest. A commitment to deliver.

Investment Philosophy and Approach

Our approach to investment management relies upon an active, unconstrained management style that is founded on the minimization of absolute risk. The objective of our investment strategy is to outperform the benchmark over a time horizon of three to five years with a lower absolute risk.

We seek to construct diversified portfolios that preserve investor’s capital over time and firmly believe that long term returns are an outcome of rigorous risk management. Our approach is based on the following core principles:

  • Not all risks can be observed from numbers alone;
  • Picking low volatile stocks is not enough, assembling them in an optimally diversified way is key;
  • To be robust, risk management needs to combine both a quantitative process and a qualitative judgment.

Unigestion’s Minimum Variance approach is based on three steps:

  • Refining the investment universe with a view to eliminating those stocks, the past price behaviour of which inaccurately reflects their current risk. At this stage, we closely monitor liquidity, the company’s financial soundness and specific risks (such as potential M&A events);
  • Using an optimizer to construct a prototype portfolio with the lowest volatility;
  • Qualitatively assessing each candidate stock for inclusion in a portfolio. In addition, we take into account sovereign credit risk to limit exposure to countries facing challenging financial conditions.

Our portfolios are then monitored on a continuous basis to identify any new sources of risk which may emerge from the stocks that we hold.

Through our 17 years of experience of designing active risk management techniques on several investment universes, we have learned that Minimum Variance is much more than a quantitative optimization. Human judgment and experience play a critical role in identifying and avoiding “bad” risks. We feel confident that our investment process protects our investors against the pitfalls of quant-only techniques.​