A completely automated optimization strategy for global minimum-variance portfolios based on a new test for structural breaks

摘要

We present a completely automated optimization strategy which combines the classicalMarkowitz mean-variance portfolio theory with a recently proposed test for structural breaks in co-variance matrices. With respect to equity portfolios, global minimum-variance optimizations, which basesolely on the covariance matrix, yield considerable results in previous studies. However, nancial assetscannot be assumed to have a constant covariance matrix over longer periods of time. Hence, we estimate the covariance matrix of the assets by respecting potential change points. The resulting approachresolves issues like timing or determining a sample for parameter estimation. Moreover, we apply theapproach to two datasets and compare the results to relevant benchmark techniques by means of anout-of-sample study. It is shown that the new approach outperforms equally weighted portfolios andplain minimum-variance portfolios on average.