Literature on allocation across multiple strategies on the same asset

[Martinghoul]

@cjbuckley4
The question you're asking is actually a very interesting one. I don't have a good objective answer for you, but I can tell you where my preferences lie. Specifically, I find it easier and more logical to look at strategies separately and then construct a portfolio out of them, rather than aggregate positions into a "blob" and manage the resulting Greeks/factors. That said, I know and have worked alongside people who choose to approach their risk differently. In my mind, this is sort of a prop vs mkt-maker dichotomy of risk management. Again, I am not sure there's a good way to decide which approach is better objectively, since a lot depends on what you're personally comfortable with.

 

[nonlinear5]

So the VaR equation looks like sigmaOfCommonAsset*(whatever number of contracts these strategies want to hold cumulatively). My concern is that VaR will jump around wildly and cause me to concentrate too much risk. I'd much prefer a method that targets a keeps the risk a lot more static.

I used to be uncomfortable with that notion of the constantly changing risk/exposure. But then it occurred to me that this is how it should be. Your total exposure *should* be proportional to the payoff probability/magnitude. There is nothing wrong with it. I run an automated system which consists of about 15 trading strategies. I am totally content when none of them are in position (i.e. when the system risk/exposure is zero). I am also totally content when multiple strategies are in position, because I know that it happened when the market presented favorable odds (i.e. multiple "setups" have occurred). Constant level of risk/exposure makes no sense, because it treats all market conditions as "same" from the portfolio allocation perspective. A better way to cope with varying risk is to simply reduce the overall leverage.