Quant funds utilize computer algorithms to guide their investment strategies. These computerized methods pick securities based on quantitatively identifiable characteristics. Rather than selecting stocks with a good story (e.g., Alibaba (BABA)), they select stocks based on various fundamental or technical criteria.
The basic idea behind quant funds is to identify anomalies or return factors that lead to higher returns or less volatility. By giving a preference to investments with these characteristics, higher returns, less volatility or both are sought. It can be an unemotional way to invest as long as personal biases are not allowed to interfere with either the creation or the execution of the model. The better you are able to stick to the model, the more you will be able to invest like a quant fund.
Creating your own model does require a level of comfort with a good screening program such as our Stock Investor Pro and, depending on the model used, a spreadsheet. If you are unwilling or unable to do the mathematical and computer work, following a quantitatively oriented screen or buying a strategic beta fund may be the better option.
If you are comfortable with the computer aspect, you then need to be aware of what characteristics are similar and what are different. For example, dividend yield and the price-earnings ratios are both valuation indicators, whereas debt-to-equity (leverage) and relative price strength (price momentum) are different measures. You want give considerable thought to what goes into your model and whether it gives the model a tilt. There is nothing wrong with having a model that emphasizes, say, value, but you should be intentional in trying to achieve that focus.
The next step is to determine how to use the various factors to influence which stocks top your results. Your choice is to use either factor or absolute criteria. The factor approach weights each criterion (or composites of like criteria) equally and then selects the stocks with the best factor scores. This was the approach discussed by Ronen Israel, a principal at asset manager AQR Capital, at Morningstar’s ETF Conference last week. It is also the approach Jim O’Shaughnessy talked about at the 2013 AAII Investor Conference. With an absolute criteria approach, you are telling the screen to identify only those stocks meeting specified criteria (e.g., a price-to-sales ratio below 1.2) and to ignore anything outside of those boundaries.
Ronan told me after his presentation that he likes the factor approach. He believes factor models do a better job of maintaining the diversification benefits of equally weighting criteria with different attributes. If absolute criteria are used, then the model can end up being biased toward one particular investing style (e.g., value).
Absolute criteria works better when you want more a style-plus strategy, such as value with a momentum tilt. This would ensure you don’t buy stocks above a certain valuation limit, while still taking advantage of momentum characteristics.
Once you decide on the appropriate approach to follow, you have to pick the specific criteria by which to identify stocks. There is no single right answer as to what to include. O’Shaughnessy’s model uses value, financial strength and earnings quality composites. Israel used an example of value, momentum and defensive (earnings quality). Research Affiliates’ RAFI Index looks at book value, cash flow, sales, and gross dividends. The key for you is to seek factors that complement each other and add benefits to the strategy.
I think it is helpful to take a step back at this point and ask whether your approach is logical. Think about what type of stocks your strategy is designed to identify. Additionally, take a look at the criteria in your model. Are they factors proven to individually lead to higher returns or reduce risk in the past? What you want to watch out for are anomalies that do not have a logical reason for working. (This is a common problem with backtesting; you may end up with odd combinations of criteria with good historical returns for no particular reason.)
Your final step is execution. You need to determine your sell rules, whether you will rebalance your holdings, and if so what amount to use to add new holdings. Keep in mind the two advantages you have over all quant funds: scale (you can buy less frequently traded stocks) and no requirement to report performance. The latter is a big advantage because all strategies will underperform for periods of time, even when their long-term returns are very impressive.
The Week Ahead
I accidently forgot to mention in last week’s email that Rosh Hashanah is this week. The Jewish holiday started last night and goes until tomorrow evening. L’Shana Tova to those who are celebrating it.
Only four members of the S&P 500 are scheduled to report earnings: Cintas Corp. (CTAS) on Monday, Walgreen Company (WAG) on Tuesday and Constellation Brands (STZ) and McCormick & Company (MKC) on Thursday.
The economic calendar has August personal spending and income and the August pending home sales being released on Monday. Tuesday will feature the July S&P Case-Shiller home price index, the September Chicago PMI and the Conference Board’s September consumer confidence survey. The September ISM manufacturing index, the September ADP employment index, the September PMI index and August construction spending will be released on Wednesday. Thursday will feature August factory orders. The September employment report (including the unemployment rate and the change in nonfarm payrolls), August international trade and the September ISM non-manufacturing index, will be released on Friday.
Two Federal Reserve officials will speak: Chicago president Charles Evans on Monday and Governor Jerome Powell on Tuesday.
The views and opinions expressed herein are the author's own, and do not necessarily reflect those of EconMatters.
© EconMatters All Rights Reserved | Facebook | Twitter | Email Subscribe | Kindle