# How do you add to a portfolio (A retarded question)

I confess I pass my CFA L3 but have no idea how to do it. Call me an idiot.

Let say I have \$60 in equity and \$20 in bond and wish to add a \$20 managed future in it. There are 5 managed futures for me to choose and I wish to evaluate them quantatively.

Should I generate 5 portfolios and use mean-variance as the criterion to evaluate? Or can I just evaluate by the sharpe ratio of the 5 managed futures? Did CFA teach this?

Ultimately, you want the final portfolio to have the highest Sharpe ratio. So, which one depends on 1) correlation with the existing (75% equity / 25% bonds) portfolio and 2) the Sharpe ratio of the managed futures.

There’s something in Level 2 where you multiply the sharpe ratio of the existing portfolio (75% equity, 25% bond) by its correlation with the managed future. If the sharpe ratio of the managed future is higher than that number, then the portfolio would benefit from having the managed future in it.

I am not 100% sure if the following is a true corrolary, but I suspect you may want to calculate the following product for each futures strategy:

(correlation with stock/bond portfolio) * (Sharpe ratio of managed futures)

And choose the highest one, PROVIDED that it passes the first criterion I mentioned.

If the managed futures aren’t correlated with each other, you also might get a better portfolio by having several managed futures in there - however, most managed futures are trend following and are likely to be highly correlated.

Thanks bchadwick. Things that I don understand but memorize are coming back to haunt me

I learn many ratios throughout my finance life. Just wondering can I say sharpe ratio is the most important as sharpe ratio is based on modern portfolio theory? I will only have to generate the 11 hypothetic portfolio and see which one has the highest sharpe ratio?

Well, you probably want to do a little stress testing too.

The Sharpe ratio may not always be the best measure for alternative portfolios if they have highly skewed returns. Managed futures are probably less skewed than portfolios that have lots of options in them.

I tend to look at the Sharpe ratio first, but then you also want to look at things like maximum drawdowns and questions like “what are the things that could make the portfolio perform worse than has ever been observed to-date.” It’s hard to come up with answers to questions like that, but it’s worth at least trying to ask.

Bchadwick is correct that in theory you want to pick the portfolio with the highest Sharpe. He is also correct about a very useful rule of thumb that says that a new component deserves an allocation if its Sharpe exceeds the product of the original portfolio Shapre and the correlation of the component and the original portfolio. You might want to be careful to give that component a 20% allocation. It’s possible that the best weight is 5% or 40% - the product doesn’t say anything about the optimal allocation to the new component.

If you are considering selecting 1 out of 5 managed futures options with target allocation of 20% and target allocation for stocks of 60% and bonds of 20%, just create a portfolio for each of the 5 options with those weights and compare their risk-adjusted return.

If you are considering multiple managed futures investments with aggregate allocation of 20%, that would be a more complex optimization problem that you can solve using Excel solver. I can help you set it up if you have difficulty doing that.

Good point, maratikus. I thought about mentioning that 20% futures allocation might not be optimal, but figured that that would add a lot of extra detail. It is worth analyzing, however, because if the futures strategy is really volatile, a 20% allocation may well be too much. A low-tech way to calm it down would be to allocate some of that 20% to cash and the remainder to the futures strategy. Of course, if there is a minimum investment size, that may not always be a feasible option.

Ideally, you’d just try to find optimal weights, for stocks, bonds and futures and see how they compare with your 60-20-20 recommendation.

Interestingly, one of my tasks this month for a client is to do something like this and explain how their new CTA strategy fits in with some of their existing strategies.

@Maratikus, I’ve been thinking about you lately and some of the strategy discussions we had a little while ago. Are you in NY any time soon?

bchadwick, I don’t plan on going to NYC for a while. I will email you soon.

I too passed CFA L3 and have no idea what you’re asking here…none zip zero…

Don’t worry, Frank. It’s just theory. It’s not for you.