Semi-Metric Portfolio Optimization: A New Algorithm Reducing Simultaneous Asset Shocks

This paper proposes a new method for financial portfolio optimization based on reducing simultaneous asset shocks across collection of assets. may be understood as an alternative approach to risk reduction in mathematical quantity. First, we apply recently introduced semi-metrics between finite sets determine the distance time series’ structural breaks. Then, build classical theory Markowitz and use this breaks our penalty function, rather than variance. Our experiments are promising: synthetic data, show that proposed does indeed diversify among series with highly similar enjoys advantages over existing metrics sets. On real illustrate performs well relative nine other commonly used options, producing second-highest returns, lowest volatility, second-lowest drawdown. The main implication management is potentially sharp associated drawdowns during periods breaks, such market crisis. adds considerable literature techniques econometrics could complement these via averaging.