Portfolio optimization : equally weighting strategies vs. index investing vs. efficient frontier portfolios : an empirical analysis

Otto, Hans-Philipp (2012-12)

Thesis (MBA)--Stellenbosch University, 2012.


This research report is conducted in the field of portfolio optimization. Regarding the existing literature this research paper is set in context of the academic discussion triggered by DeMiguel, Garlappi and Uppal (2009) concerning the perfomance of the naïve investment strategy in comparison to optimized portfolios and extended by the indexing approach. Therefore, it investigates on the question whether the naïve investment strategies outperform the strategy of index investing as well as the minimum and mean variance portfolios in the investment horizon of the EURO STOXX 50 in the timeframe from 03.01.2003 to 02.07.2010. Outperforming is defined via the following measurements, namely return, variance, Sharpe ratio, value at risk, certainty equivalent return and turnover rate. In addition, modifications of the investment strategies are applied such as the rebalancing of the naïve investment strategy and different scenarios are included such as the consideration of transaction costs and costs of index investing as well as the usage of two different data frequencies in order to conduct the robustness test. The two main measurements Sharpe ratio and value at risk are verified regarding their explanatory power by the usage of the robust inference method for the bootstrapping of the Sharpe ratio and the Jarque-Bera test for the normal distribution required for the value at risk measurement. The research in this paper is conducted through MATLAB which is a numerical computing environment and fourth-generation programming language. The aggregated outcome of this research paper in regard to the respective timeframe and investment horizon is that in the main scenario which is based on weekly input data the minimum variance investment strategy outperforms all other investment strategies consistently in all measurements except for the turnover which is compensated by consistent results in case of inclusion of transaction costs and costs of index investing. Furthermore, the rebalanced naïve investment strategy and the index investing strategy share the second place with a slight advantage in the overall perspective for the rebalanced naïve investment strategy as it dominates the index investing strategy in regard of return, Sharpe ratio and certainty equivalent return while it is only outranked by the index investing strategy in the risk related measurements variance and value at risk. All other investment strategies underperform their peers.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/95621
This item appears in the following collections: