The Old Mutual Asset Managers Prize goes to the best portfolio produced under the specification of Assignment 2 in CF963, "Learning and Computational Intelligence in Economics and Finance", 2011-12.
Thanks to everyone who submitted your portfolio. The assignment accounts for 15% of the module, but you have clearly spent more time than this percentage suggests.
Many of the submissions are innovative. Many have chosen to pick five stocks first, and then decide what their weights should be. Heuristics have been used to pick the five stocks. Some have pointed out that to choose five out of ten stocks, there are 252 possible combinations; exhaustive search of these 252 stocks is feasible. That is quite a promising approach for the given problem. After picking the five stocks, determining the weights is the next challenge. Various techniques have been tried, including Mean-Variance, Lagrange Algorithm, Genetic Algorithm.
Before announcing the best portfolio, I must note that judging is not as straight-forward as one might think. Different submissions compute different Sharpe Ratios for the same portfolio. This may be due to differences in rounding or book-keeping. In the end, I judged the portfolios with my own spreadsheet, which I have carefully checked (no guarantee that it is 100% correct). I have also cross-checked portfolios with some of the submissions.
The Old Mutual Asset Managers Prize goes to Riccardo Lattanzi (photo above, right), whose portfolio is (VOD 30.2261%, BATS 69.7654%). Its Sharpe Ratio is 76.4864, according to my spreadsheet. Riccardo Lattanzi used Microsoft Excel's "Solver" (which is an add-in) to find weights for the five stocks that he chose heuristically and scientifically.
The Prize comprises a certificate (shown here), a bottle of wine (donated personally by Dr Alentorn, Old Mutual Asset Managers) and a visit to Old Mutual (expenses paid for) to gain an insight of how they work (pictured above). Riccardo Lattanzi will conduct an MSc project based on portfolio optimization with Dr Alentorn.