So what exactly is quantitative investing and why do I think it is potentially preferable to the traditional method of analyzing company research reports, accounting statements, and meeting with company management? A traditional manager digs very deeply into each investment. Because the portfolio he or she is going to hold is very concentrated (holding maybe 10 to 25 stocks), the level of investment conviction has to be very, very high. That is the case because the risk of ruin if a stock “blows up” in a 10-stock portfolio is very real.
Quant investors on the other hand have the ability to study thousands of stocks at one time. And in fact, they are looking at much of the same data (factors) that traditional investors analyze. A quant investor is concerned about valuation, recent price trends, cash levels on the balance sheet, earnings revisions and the quality of earnings. But quant portfolios take many more, smaller bets and as a result only have to be right more than half the time.
Quant models are a great source of idea generation, but in this day and age I would argue that you can’t be a “pure quant” and succeed. This is due in part to the fact that price, earnings, and valuation data are so readily accessible in the age of computers, online brokerage, and free data services such as Yahoo! Finance. Information in its pure form has become a lot more difficult to exploit.
So how can quant managers still potentially generate alpha in this new world of data proliferation? The key is to focus on areas of the market that are still very inefficient. Proponents of efficient market theory will say that everything is already priced into the market. But there is indeed alpha to be captured out there! In order to gain an advantage relative to passive management, you need to focus on picking stocks that are misunderstood, either underappreciated or overappreciated, or just plain not followed or owned. Misinformation creates mispricing opportunities.
So the lower the coverage and smaller the market capitalization, the easier it is to potentially generate alpha. That is why I focus my investment efforts on micro cap (under $1 billion), small cap (under $3 billion), and mid cap (under $15 billion). These market capitalization segments offer much better ponds from which to fish. But companies that have lower coverage and smaller market capitalization generally also carry with them more risk. That’s where human intervention comes in. You can’t just “follow the model” blindly. It is important to conduct extensive fundamental research on each of the companies and really understand the data inputs and the story behind the numbers. Portfolio construction is also a critical component of quant investing, particularly when investing in smaller capitalization companies. It is very important to hold a diversified portfolio by sector, industry, market capitalization, and style.
One of the elements that quant investing can exploit is human behavior. Investors tend to anchor on the past or be overconfident about the future. They get caught up in the glamour aspects of companies or hold grudges against them for past failings. Quant models do not fall in love with companies or refuse to ever own them again. Even seasoned investors make investing mistakes driven by emotion and not facts. The key to investing success is a rational investment approach coupled with informed human insight. This approach can deliver the most consistent and repeatable results over time.
Disclaimer: Past performance is no guarantee of future results.
The post The case for quant investing appeared first on Smarter Investing
Covestor Ltd. is a registered investment advisor. Covestor licenses investment strategies from its Model Managers to establish investment models. The commentary here is provided as general and impersonal information and should not be construed as recommendations or advice. Information from Model Managers and third-party sources deemed to be reliable but not guaranteed. Past performance is no guarantee of future results. Transaction histories for Covestor models available upon request. Additional important disclosures available at http://site.covestor.com/help/disclosures.