In this paper, I evaluate the out-of-sample performance of the portfolio optimizer relative to the naïve and market strategy on the Swedish stock market from January 1998 to December 2012. Recent studies suggest that simpler strategies, such as the naïve strategy, outperforms optimized strategies and that they should be implemented in the absence of better estimation models. Of the 12 strategies I evaluate, 11 of them significantly outperform both benchmark strategies in terms of Sharpe ratio. I find that the no-short-sales constrained minimum-variance strategy is preferred over the mean-variance strategy, and that the historical sample estimator creates better minimum-variance portfolios than the single-factor model and the three-factor model. My results suggest that there are considerable gains to optimization in terms of risk reduction and return in the context of portfolio selection.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:uu-218024 |
| Date | January 2014 |
| Creators | Ramilton, Alan |
| Publisher | Uppsala universitet, Företagsekonomiska institutionen |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.003 seconds