Portfolio performance attribution

In this paper, we provide further insight into the performance attribution by development of statistical models based on minimizing ETL performance risk with additional constraints on Asset Allocation (AA), Selection Effect (SE), and Total Expected Value Added by the portfolio managers (S). We analyze daily returns of 30 stocks traded on the German Stock Exchange and included in the DAX30-index. The benchmark portfolio is the equally weighted portfolio of DAX30-stocks. The portfolio optimization is based on minimizing the downside movement of the DAX-portfolio from the benchmark subject to constraints on AA, SE and S. We investigate also the distributional properties of AA, SE and S sequences by testing the Gaussian distribution hypothesis versus stable Paretian hypothesis. Finally, we propose an empirical comparison among suggested portfolio choice models comparing the final wealth, expected total realized return of the optimal portfolio, and performance ratios for obtained sequences of excess returns.