Quantitative Investing: Introduction to data-driven investing and quantitative investment strategies

Understanding the difference between quantitative and fundamental investing

The field of quantitative equity investing, which only emerged in the 1980s and 1990s, is now an established part of the asset management industry. While many areas of the financial services industry are being disrupted by technology and new distribution channels, quant investing continues to flourish.

Quantitative Investing
Image Source: whiteMocca / Shutterstock.com

Quantitative portfolio management, which is based on empirical evidence, eliminates the negative effects of emotion on decision making, is cheaper than fundamental analysis, and allows small teams to cover a large universe of securities. The potential of quant investing is still to be fully realised, with several new advances under way and on the horizon. This article will give you an introduction to quantitative investing.

What is quantitative investing?

Trading Room / Stock Quotes
Image Source: ViblyPhoto / Shutterstock.com

Quantitative investment management makes use of statistical and mathematical models to study the behaviour of stocks, as well as other asset classes. There are two distinct parts to quantitative investing; research and implementation. Research may be based on proprietary research or by using published academic papers.

The research is used to construct a model that identifies stocks with a higher than average probability of outperforming a benchmark index. To implement a model, stocks will typically be assigned a score based on one or more characteristics (or factors) and then ranked. A quantitative investing portfolio will typically hold the top ranked stocks, and then be rebalanced at regular intervals or when it is out of line with a model. Quantitative techniques can be used to manage both long only, and long / short portfolios.

Why quantitative investing?

Bloomberg Terminals
Image Source: Rokas Tenys / Shutterstock.com

When an active asset manager makes an investment decision, it will usually be based on how they believe the company will perform in the future, with the assumption that strong company performance would lead to strong share price performance. These decisions are based on subjective analysis of the company’s management and products, and the market and economic environment it is operating in.

Actively managed funds have been benchmarked against indexes since the 1960s. Over time it has become clear that the majority of actively managed funds do not outperform their benchmark indexes consistently. Advances in technology during the 1970s meant that by the early 1980s investment analysts could for the first-time study very large data sets. This quantitative analysis allowed investors to find out which types of stocks outperformed over time.

Quantitative investing made three things possible – studying larger numbers of stocks simultaneously, decisions based on empirical evidence rather than on subjective forecasts, and a systematic approach to portfolio management. Early research determined that certain anomalies existed to explain stock price performance. Value, momentum and market value were the first factors found to lead to outperformance. Over time, other factors and combinations of factors have been found to lead to outperformance.

Quantitative investment analysis is also useful for asset allocation and risk management. It allows a portfolio to be constructed or analysed based on long term expected returns and volatility. This allows portfolios to be created to suit the individual needs of investors. These days most funds use a quantitative approach for at least some aspect of their portfolio management process. Even if it is not used for stock picking, it will usually be used for risk management or asset allocation.

Quantitative vs. fundamental investing

Quantitative vs. Fundamental Investing
Image Source: Phongphan / Shutterstock.com

The more traditional active and fundamental investment approaches are usually based on bottom up analysis and on forecasts of company earnings and economic growth. Fundamental analysis also looks at qualitative factors like the quality of management and the strength of the balance sheet. When using quantitative factors in capital investment decisions, portfolio managers are looking for the factors that have proven to reliably lead to outperformance. Rather than making investments based on subjective forecasts, they are made based on empirical evidence.

Quantitative investing models are based on probabilities and an expected distribution of returns. This means the expected risk and return can be more accurately predicted, but this also requires a large enough sample size to be effective. Quant funds therefore typically hold a higher number of securities than actively managed funds.

Investment decisions for an actively managed fund are made by the fund manager with a large amount of discretion. For quant funds, buy and sell decisions are made by a model, with very little room for discretion on the part of the fund manager.

Types of quantitative investment strategies

Stock Trader
Image Source: g-stockstudio / Shutterstock.com

While most quantitative investment models overlap one another, and may also have their own unique features, most strategies have elements of a few of the following strategies:

Factor investing models are used to select stocks that share one or more characteristics that have historically led to outperformance. General factors include value, momentum, market capitalization, and growth. More specific factors include ratios like price to book, price to free cash flow and return on equity. Quantitative investing factor models usually score each stock on a range of metrics and then calculate a total score which is used to rank stocks.

Event-driven arbitrage strategies take advantage of price patterns that typically occur before or after events. Events include earnings releases, economics data announcements, corporate action, and regulatory changes. Portfolios are constructed by buying or selling short securities to lock in profits if price action follows a typical pattern.

Systematic global macro strategies are based on quantitative analysis of the economies in each country and region. This analysis is used to allocate capital to countries, regions, asset classes and sectors that have favourable fundamentals.

Global Economy
Image Source: welcomia / Shutterstock.com

Risk parity funds balance a portfolio’s risk across asset classes, based on how each asset class tends to behave in different types of environments. The idea is that volatility and losses in one asset class will always be offset by the other asset classes. This approach doesn’t necessarily outperform equity only funds but can produce better risk adjusted returns over time.

Statistical arbitrage is a one of the more active quant trading strategies. This is a mean reversion approach that looks for mispricing based on the relationships between securities. Long and short positions are opened in related stocks to profit when prices revert to normal. Statistical arbitrage also makes use of financial ratios to identify mispriced assets.

Managed futures, also known as CTAs, commodity trading advisors and trend following hedge funds, use a systematic method to follow major market trends. Traditionally these funds have focussed on futures markets, but increasingly they are also active in the stock market.

Adjusted Risk / Reward Ratio
Image Source: Olivier Le Moal / Shutterstock.com

Smart beta strategies are used to systematically manage passive investing vehicles like ETFs and mutual funds. Rather than using market capitalization to weight stocks, other factors can be used to improve the risk adjusted return of a portfolio.

Quantitative value funds use a methodical approach to go through each line of the income statement and balance sheet of each company. An aggregated value score is then calculated and used to rank stocks. This systematic value investing approach can be very effective, but a long-term time horizon is required.

A.I. and Big Data based strategies are the newest type of quant strategy. They attempt to find new sources of alpha using techniques and data that until recently had not been used in the fund management industry.

Benefits of quant strategies

Rational Trading Decisions
Image Source: Olivier Le Moal / Shutterstock.com

Because quant trading decisions are made by a computer model they are not influenced by human emotion. When people make investment decisions they are often influenced by fear or greed. This applies to both entering and exiting positions, where discipline is often a problem for investors. Furthermore, quantitative investing can actually take advantage of irrational decision making in the market place.

Small teams of quantitative analysts can cover a very large number of securities. They can cover multiple sectors, regions and countries, without having to hire new analysts. Quant teams therefore have more opportunity to find securities likely to outperform. It also means analysis is cheaper on a “per stock” basis.

Quantitative investing is evidence based, meaning the results are more predictable, especially in terms of the expected risk and return profile. They can therefore be better aligned with the needs of different investors. Once created, quantitative models can easily and cheaply be tested on different markets, with or without modifications being made.

Disadvantages of quant strategies

Failing Robot / Quant Machine
Image Source: FGC / Shutterstock.com

Because quant strategies are based on an expected return distribution and probabilities, a fairly large number of holdings is required. This can lead to dilution of returns. Quant strategies typically require long time periods to perform and will often underperform their benchmark over shorter time horizons. This is not true for all quant funds, and new data sources are now being used to create models that generate alpha in the short term. Most quantitative funds are unable to take subjective factors into account.

While funds with a strong momentum bias can capture the performance of growth stocks, funds based on other factors will usually miss out on high growth stocks. Quant strategies are vulnerable to sudden increases in volatility and flash crashes that can be created by other algorithmic trading strategies. The fact that quantitative funds are managed without any discretion can be a double-edged sword. In most cases the dispassionate nature of decision making is an advantage, but there are occasions where it can be a disadvantage.

Quantitative investing today

Quantitative Investing through Big Data Analysis and Artificial Intelligence
Image Source: Elnur / Shutterstock.com

Today, Wall Street has embraced quant investing, and quantitative techniques are used to manage most types of investment funds, including mutual funds, hedge funds, ETFs and segregated portfolios. Quantitative techniques are also used for asset allocation and risk management and to align portfolios with the needs of clients.

The new frontier for quant investment is the creation of strategies that fully embrace technology. Artificial intelligence is being used to find more obscure patterns and relationships between asset prices and data from other data sources. Big data is being used to source and mine new sources of data that may lead to alpha generating ideas. User generated data is being used to measure investor sentiment which can be mapped to asset prices.

Quant platforms like Quantopian and Quandl are also being used to crowd source ideas and allow quant analysts to collaborate and source data. Robo advisors, which allow individual investors to invest or save for retirement or specific events, make use of quantitative models to allocate capital. And lastly, social trading platforms allow traders to verify their performance and manage funds for individual investors.

The future of quantitative investing

Future of Quantitative Investing
Image Source: Gorodenkoff / Shutterstock.com

Quantitative investing is advancing on several different fronts, and the future is likely to see the convergence of these different techniques and platforms. Developments in other areas, including the introduction of new investment products and asset classes (cryptocurrencies and tokenised securities for example) will create new opportunities.

The continued globalization of markets will also play a role in the future, as investors will be able to access new markets. The biggest opportunities may lie with A.I. and Big Data. These technologies, when used together, allow analysts to find relationships between stock prices and data not traditionally used by investors. Satellite images, social media content and GPS data from vehicles and devices are all potential sources of intelligence.

Sentiment is a factor that is growing in importance for quantitative investing. Both A.I. and Big Data are being used extensively to model sentiment and its predictive powers. Advances in A.I. may eventually allow qualitative factors to be modelled. This would close the gap between quantitative and traditional active fund management, by taking more subjective factors into account.

The industry is also likely to become more competitive. Only the quant asset management firms committed to evolving and finding new quantitative strategies will be able to consistently generate alpha, and ultimately survive.

Conclusion: Quantitative analysis as systematic approach to investing

Quantitative analysis has introduced a more scientific and systematic approach to investing. There are several advantages to making investment decisions based on empirical evidence, including lower relative costs and the elimination of emotion from decision making. 

A strategy based on a quant model is not a silver bullet, and there is no guarantee of performance, but for the most part quant funds have a better chance of achieving their objectives. The recent introduction of new products, technologies and asset classes suggest that there is still a long way to go and the industry will continue to grow and evolve in the coming decade.

About Richard Bowman
Richard Bowman is a writer at Catana Capital, analyst and investor based in Cape Town, South Africa. He has over 18 years’ experience in asset management, stockbroking, financial media and systematic trading. Richard combines fundamental, quantitative and technical analysis with a dash of common sense.

Be the first to comment

Leave a Reply

Your email address will not be published.


*