Have you ever wondered how most professional equity managers create a portfolio? I'm not talking about picking stocks, I'm talking about allocating your capital into a buy list so that the portfolio is 100% invested. As it would happen, I used to trade and manage billions in equity for countless institutional equity funds and do you know how many equal weight portfolios I have seen? One. The reason is that professional fund managers add value in two ways: stock selection and portfolio construction. And equal weighting a portfolio isn't adding any value. So if you are doing that or just ad-hoc weighting your portfolio, it's time to take your investing to the next level.
In order to execute this advanced strategy we will need to leverage some sophisticated mathematics behind the scenes. For this article we will use the free portfolio analytic web application called BlairQA (blairqa.com), of which I am a core developer. Of course you can also use any of the paid subscription based services like Barra, Northfield, Axioma, and Bloomberg as well.
How do we know if our portfolio allocation is good or not? The answer is by using Information Ratio [IR]. The IR captures the two most important characteristics of your portfolio: return and risk. The formula is simple: IR = return / risk. This makes a lot of intuitive sense because in finance we want return and the cost of that return is risk. This is the central concept in all of modern portfolio theory. If we want more return we must take on greater risk. This is where the optimizer will come in. An optimizer will create the lowest risk portfolio for any desired level of return, thereby maximizing the conditional information ratio. It sounds complicated but it's not. For example, say we have a portfolio of 5 stocks and we desire a 20% expected return. There are many ways we could weight the portfolio to achieve that 20%, but there is exactly one weighting that is optimal, i.e. takes on the least amount of risk to get the 20%. And that is how you get a free lunch. You take the same stocks you already own, and change the weights to get more return (or the same amount of return) for less risk.
Here is a real world example to illustrate how you can use an optimizer to maximize your information ratio. I created a portfolio in BlairQA that holds five stocks: Netflix (NASDAQ:NFLX), Lumber Liquidators (NYSE:LL), Buffalo Wild Wings (NASDAQ:BWLD), Tesla Motors (NASDAQ:TSLA), and Wynn Resorts (NASDAQ:WYNN). I assigned ad-hoc weights and set each stock's expected return equal to it's trailing 12 month return. Setting expected returns is a different article but you could also use analyst price targets or your own forecasting model.
We ran the optimizer and configured it to target the same expected return as our initial portfolio. So in effect we said "Give us the least risky portfolio that has the same expected return.". This will leave the IR numerator (expected return) unchanged but should lower the denominator (risk) thereby increasing our portfolio IR.
As you can see, the weights have all been changed. The portfolio is still 100% invested and the expected return is that same as the initial portfolio, 90.155. But take a look a the Portfolio Risk to Cash, our initial portfolio had risk of 24.55 while our new optimized portfolio has risk of only 22.898. As expected, because the risk has decreased while our expected return has stayed the same, our IR has increased from 3.67 to 3.94. This allocation is optimal at the 90.155 level of expected return. To put it another way, there is no other portfolio allocation with that level of expected return that has lower risk.
We can even validate that the optimizer results are accurate by testing the historical volatility of the new optimized portfolio versus our initial portfolio. The optimal portfolio is in light blue and you can see that it has lower risk at almost every point in time over the past year. As an aside, we didn't need our initial portfolio to start with any weights. For example we could have created a portfolio with the same stocks and assigned weights of zero to each. The optimal result will be exactly the same.
Information Ratio is the metric that professional portfolio managers use to evaluate portfolio allocations. It is a powerful ratio that easily quantifies the two essential portfolio characteristics: return and risk. By always looking to maximize your information ratio for a given level of desired return you will never 'overpay' for return giving yourself the best statistical opportunity to realize alpha.
Disclosure: I am long LL, BWLD, NFLX, TSLA, WYNN. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.