Greetings all -- There are several ways to create a portfolio from a group of individual tradable issues or systems.
The Markowitz CAPM Mean-Variance model has traditionally been applied to a portfolio of individual equities, assuming that they will be bought and held. You could treat each trading system as an equity, compute its mean and variance, and plot it as a point on the M-V graph. Then apply CAPM methods to select the optimal portfolio -- select the issues that give the best return for the least variance, given the portfolio's objective. But if you are dealing with a number of trading systems, each of which has already has an equity curve created by the active trading of one or more issues, you might prefer to align the daily equity of the individual systems by date, then search for the weightings that give the best, according to your objective function, portfolio return. Of course, you could apply the same technique to the daily closing prices of the equities. In the first example -- the CAPM model -- you will never choose a system that has a negative return as a component of the portfolio. In the second example, you might. Thanks, Howard
