<I guess the real question is, what is the optimum
allocation given you have an appropriate number of high quality growth companies?>
Bruce:
If you can provide me with what you think would be the most appropriate return/probability profile for QCOM, SNDK, and WIND (you may disagree with what I have used), I would be very happy to run some numbers for you, in order to get an idea of the optimum allocation. As you can gather, the return/probability profiles are critical to the result, as well as your own utility function.
On the question of independence, random numbers are generated independently for each stock for each of the 600 or so simulations for each allocation tested. Mathematically, this means that the performance of each stock should be totally independent of each other. In practice, this is not going to happen, as all the stocks are going to be affected by the general market movements and by general movements in particular industries, such as high-tech. However, there is a large element of crystall-ball gazing when we develop our return/probability profiles, and given the high likelihood that we may be wrong, I would not worry too much about independence at this stage, as long as the companies' financial positions are not too directly connected with each other (such as might happen is one is the sole supplier of parts for the other).
Erwin |
