Quantitative analysis: A simple overview

All of the potential highs, lows, and sentiments associated with investing can overshadow the ultimate goal: making money. In an effort to focus on the latter and eliminate the former, the “quantitative” approach to investing seeks to pay attention to the numbers instead of the intangibles.

Harry Markowitz is generally credited with beginning the quantitative investment movement when he published a “Portfolio Selection” in the Journal of Finance in March of 1952. Markowitz used math to quantify diversification, and is cited as an early adopter of the concept that mathematical models could be applied to investing.

Robert Merton, a pioneer in modern financial theory, won a Nobel Prize for his work research into mathematical methods for pricing derivatives. The work of Markowitz and Merton laid the foundation for the quantitative (quant) approach to investing.

Unlike traditional qualitative investment analysts, quants don’t visit companies, meet the management teams or research the products the firms sell in an effort to identify a competitive edge. They often don’t know or care about the qualitative aspects of the companies they invest in, relying purely on math to make investment decisions.

Hedge fund managers embraced the methodology and advances in computing technology that further advanced the field, as complex algorithms could be calculated in the blink of eye. The field flourished during the dotcom boom and bust, as quants largely avoided the frenzy of the tech bust and market crash.

While they stumbled in the Great Recession, quant strategies remain in use today and have gained notable attention for their role in high-frequency trading (HFT) that relies on math to make trading decisions. Quantitative investing is also widely practiced both as a stand-alone discipline and in conjunction with traditional qualitative analysis for both return enhancement and risk mitigation.

The rise of the computer era made it possible to crunch enormous volumes of data in extraordinarily short periods of time. This has led to increasingly complex quantitative trading strategies, as traders seek to identify consistent patterns, model those patterns and use them to predict price movements in securities.

The quants implement their strategies using publicly available data. The identification of patterns enables them to set up automatic triggers to buy or sell securities.

For example, a trading strategy based on trading volume patterns may have identified a correlation between trading volume and prices. So if the trading volume on a particular stock rises when the stock’s price hits $25 per share and drops when the price hits $30, a quant might set up an automatic buy at $25.50 and automatic sell at $29.50.

Similar strategies can be based on earnings, earnings forecasts, earnings surprises and a host of other factors. In each case, pure quant traders don’t care about the company’s sales prospects, management team, product quality or any other aspect of its business. They are placing their orders to buy and sell based strictly on the numbers accounted for in the patterns they have identified.

Quantitative analysis can be used to identify patterns that may lend themselves to profitable security trades, but that isn’t its only value. While making money is a goal every investor can understand, quantitative analysis can also be used to reduce risk.

The pursuit of so called “risk-adjusted returns” involves comparing risk measures such as alpha, beta, r-squared, standard deviation and the Sharpe ratio in order to identify the investment that will deliver the highest level of return for the given level of risk. The idea is that investors should take no more risk than is necessary to achieve their targeted level of return.

So, if the data reveals that two investments are likely to generate similar returns, but that one will be significantly more volatile in terms of up and down price swings, the quants (and common sense) would recommend the less risky investment. Again, the quants do not care about who manages the investment, what its balance sheet looks like, what product helps it earn money or any other qualitative factor. They focus entirely on the numbers and choose the investment that (mathematically speaking) offers the lowest level of risk.

Risk-parity portfolios are an example of quant-based strategies in action. The basic concept involves making asset allocation decisions based on market volatility. When volatility declines, the level of risk taking in the portfolio goes up. When volatility increases, the level of risk taking in the portfolio goes down.

To make the example a little more realistic, consider a portfolio that divides its assets between cash and an S&P 500 index fund. Using the Chicago Board Options Exchange Volatility Index (VIX) as a proxy for stock market volatility, when volatility rises, our hypothetical portfolio would shift its assets toward cash. When volatility declines, our portfolio would shift assets to the S&P 500 index fund. Models can be significantly more complex than the one we reference here, perhaps including stocks, bonds, commodities, currencies, and other investments, but the concept remains the same.

Quant trading is a dispassionate decision making process. The patterns and numbers are all that matter. It is an effective buy/sell discipline, as can be executed consistently, unhindered by the emotion that is often associated with financial decisions.

It is also a cost-effective strategy. Since computers do the work, firms that rely on quant strategies do not need to hire large, expensive teams of analysts and portfolio managers. Nor do they need to travel around the country or the world inspecting companies and meeting with management in order to assess potential investments. They simply use computers to analyze the data and execute the trades.

“Lies, damn lies and statistics” is a quote often used to describe the myriad of ways in data can be manipulated. While quantitative analysts seek to identify patterns, the process is by no means fool-proof. The analysis involves culling through vast amounts of data. Choosing the right data is by no means a guarantee, just as patterns that appear to suggest certain outcomes may work perfectly until they don’t. Even when a pattern appears to work, validating the patterns can be a challenge. As every investor knows, there are no sure bets.

Inflection points, such as the stock market downturn of 2008-09, can be tough on these strategies, as patterns can change suddenly. It’s also important to remember that data doesn’t always tell the whole story. Humans can see a scandal or management change as it is developing, while a purely mathematical approach cannot necessarily do so. Also, a strategy becomes less effective as an increasing number of investors attempt to employ it. Patterns that work will become less effective as more and more investors try to profit from it.

Many investment strategies use a blend of both quantitative and qualitative strategies. They use quant strategies to identify potential investments and then use qualitative analysis to take their research efforts to the next level in identifying the final investment.

They may also use qualitative insight to select investments and quant data for risk management. While both quantitative and qualitative investment strategies have their proponents and their critics, the strategies do not need to be mutually exclusive.