TrackRisk™: Portfolio Simulation System

Screen TrackRisk
In 2002, the founders of Gavekal-IS launched the first version of TrackRisk, a global portfolio simulation tool, combining their expertise in money management, academic research on risk, and artificial intelligence.
TrackRisk has therefore been designed by and for portfolio managers over circa twenty years.
The aim is (i) to reduce the uncertainty for money managers and (ii) to help them frame their investment decision rules in changing economic environments with the support of innovative computerized techniques.

The application addresses three areas of global portfolio management:

Phase 1

Asset Selection

Phase 2

Strategic Asset Allocation

Phase 3

Tactical Asset Allocation and Risk Management

Asset Selection

Idiosyncratic risks 

What is the historical return and volatility of the asset?
Is the asset distribution skewed towards negative or positive returns in volatile markets?
What is the risk of loss at any probability level? Minimum loss (VaR), average loss (expected shortfall), catastrophic loss (extreme event)?
What is the probability of a series of losses leading to significant drawdowns?
What is the confidence level on all calculations (Monte Carlo simulation)?

Dependency risks

What are the risk factors or the assets classes correlated with the asset?
When a risk factor declines or rises by say 1% or 2%, what is the asset’s behavior?
When a risk factor drops or rises by a much greater percentage, does the asset leverage or attenuate the shocks?

Stress tests

What was the asset’s behavior during the major market stress periods in the last 10, 20 or 30 years?


Can we replicate the investment with standard and liquid assets or derivatives?
Is the replication stable over time?

Alpha generation

Does the asset provide excess return?
If so, where does the excess return come from, and is it stable?

Follow up

Is the selected asset still the best among its peers?

Strategic Asset Allocation

TrackRisk understands strategic asset allocation, from a quantitative standpoint, as a long-term portfolio optimization under the following constraints:

  • Minimum and maximum exposure by asset class
  • Expected risk in terms of volatility, VaR, or drawdown
  • Diversification target
  • Expected returns

Portfolio optimization

TrackRisk defines a “Portfolio Scoring Function”, based on the selected constraints and optimizes allocations.
The optimization model is a genetic algorithm imitating the natural selection process of living organisms and converging towards the “best of breed” portfolio.

What if

The “What If” function is designed to try and test manually potential reallocations. It provides information on all the underlying assets of the portfolio:

  • Asset profile
  • Return/ risk/ ratios
  • Contribution to global risk
  • Dependency to risk factors
  • Stress test
White town

Tactical Asset Allocation and Risk Management

Adjust for macro conditions

What are the favorable and unfavorable economic scenarios for the portfolio?
How can we better optimize the portfolio given the current macro environment?

Test dynamic allocation strategies

What happens if exposures are switched from aggressive allocations to defensive allocations on a given scenario-when inflation is rising? when leading indicators are rolling over etc.?

Prepare for situations of market stress

What should the optimized portfolio look like in the case of a bond market sell-off? In case of an inflationary shock on oil etc.?

Data update and reporting

Data Update

The TrackRisk database is automatically updated from Bloomberg™ or Macrobond™, Excel™, or any internal or external data provider upon request, or at any predefined time and periodicity.


The quantitative analysis of TrackRisk can be exported to predefined and customizable report templates.
Screen Loading