Advances in Cointegration and Subset Correlation Hedging Methods
Hedging can be interpreted as a generalization of the portfolio optimization problem with applications in areas as diverse as trading, risk management and asset allocation. Like in portfolio optimization, one of the problems of hedging is that often one position or a subset (combination) of positions dominate the overall portfolio. A typical example is minimum-variance optimization, which may allocate most of the risk to a single holding. Equal-risk optimization also presents issues, because although the risk is equally distributed among holdings, combinations of holdings can still concentrate most of the risk (e.g., a portfolio of 5 stocks and 1 bond).
This paper solves these pitfalls by computing the Maeloc spread, i.e. the vector of weightings such that no particular holding or combination of holdings dominates the portfolio. This approach, called Mini-Max Subset Correlation (MMSC), means that not only the risk is balanced, but also the diversification potential is maximized. Empirical studies show that MMSC is related to PCA, with the advantage that MMSC estimators are more efficient than PCA hedges, and therefore should be preferred.
NOTE I now post my TRADING ALERTS into my personal FACEBOOK ACCOUNT and TWITTER. Don't worry as I don't post stupid cat videos or what I eat!