Semi-metric portfolio optimisation: a new algorithm reducing simultaneous asset shocks
share
This paper proposes a new method for financial portfolio optimisation based on reducing simultaneous asset shocks across a portfolio of assets. We adopt the new semi-metrics of <cit.> to determine the distance between two time series' structural breaks. We build on the optimal portfolio theory of <cit.>, but utilize distance between asset structural breaks, rather than portfolio variance, as our penalty function. Our experiments are promising: on synthetic data, they indicate that our proposed method does indeed diversify among time series with highly similar structural breaks. On real data, experiments illustrate that our proposed optimisation method produces higher risk-adjusted returns than mean variance portfolio optimisation. The predictive distribution is superior in every measure, producing a higher mean, lower standard deviation and less kurtosis. The main implication for this method in portfolio management is reducing simultaneous asset shocks and potentially sharp associated drawdowns, during periods of highly similar structural breaks, such as a market crisis.
READ FULL TEXT