GA Asset Management optimizes the exposure to interest rate risk in bond portfolios utilizing tactical duration management. The duration is actively controlled by taking short positions or long and short positions in the respective exchange-traded bond futures. The position to be taken is determined using quantitative methods only. That allows for different objectives to be implemented, such as taking hedge positions or maximizing yields. A duration overlay in this form can substantially improve the risk-reward profile of a bond portfolio, particularly during difficult phases of the fixed-income markets.

GA Asset Management utilizes a proprietary self-developed model to actively manage and optimize duration, which is based on the novel technology of support vector machines. Support vector machines have been successfully applied in a number of areas and in contrast to alternative quantitative models stand out due to their learning behaviour and generalization properties. They allow to greatly reduce the risk of overfitting and to detect subtle, even non-linear relationships and patterns in the data. In order to predict an optimal duration, our quantitative model uses a broad range of market factors, such as price data, the term structure and macro-economic data, and it also allows to take customer risk limits into account (see Risk Management), e.g. by providing auxiliary constraints such as the maximal conditional value-at-risk (CVaR).