We consume data. A lot of it. Naturally, it’s tough to estimate the exact amount. However, one recent estimate said around 18 zettabytes of data had been created as of 2018, and that number would increase to 175 zettabytes by 2025. One zettabyte is equal to one billion terabytes; that is big data in a nutshell.
And in a transaction-rich environment such as the financial markets, big data has the potential to change the way people operate. Although the financial industry has been slow to adopt it, big data can help analyze and interpret trends and other pieces of information that couldn’t be quantified before.
Big data can be used in conjunction with machine learning and predictive analytics to make predictions that were unrealistic or impossible in the past. Although it still requires human interaction, big data can help inform trading decisions in all-new ways.
The use of unstructured data is one of the key changes possible thanks to big data. Structured data that can be quantified and stored easily. In the context of investing, we can imagine information such as the historical share price of a stock.
Conversely, unstructured data is not so easy to quantify. It might come in the form of speech, language, or images. In the past, these things could not be quantified. But, as investors know all too well, investor confidence has big implications for the price of a company’s stock.
More broadly, unstructured data will allow us to gather data from more sources which can create a more complete picture of how a company is doing. This might include value and momentum. As it stands today, many investors have trouble understanding the true value of companies like Tesla (TSLA)–big data can help paint a clearer picture.
Technologies such as natural language processing (NLP) can help us extract this sentiment from news articles and other texts. Among other functions, NLP is a technology that helps computers understand the sentiment in text.
Google recently implemented NLP. Before that implementation, if you were to search “cats are cute,” you would get pages containing that text, but not ones containing “cats are adorable.” Thanks to NLP, computers can now understand that the meaning of both expressions is essentially the same.
Thus, NLP can be used along with big data to better understand overall confidence (or lack thereof) in a stock. It may allow analysts to quickly gauge companies at a high level instead of having to sift through individual news articles.
Using Analytics to Make Better Investing Decisions
Big data has the potential to enable better investing decisions. For investment managers, selecting investments has always been about the data. However, big data will allow them to work with many more sources of data in a more sophisticated way.
For instance, most investing decisions in the past have relied on just that–the past. They have relied on past (and current) data such as changes in the stock price and earnings per share. But as most investors know, past performance doesn’t guarantee future performance. However, big data combined with machine learning could enable us to create predictive models that better interpret past data. According to CME Group:
“New modeling capabilities linked to the big data approach, such as predictive analytics and machine learning, could change the nature of investment research by creating models that “think” and are able to draw forward-looking conclusions. This could lead to a convergence of quantitative fundamental models that focus on value with systematic trading programs that focus on price. The result could be a new type of automated portfolio management that focuses on “future value” and acts on “likely” events that may not have yet occurred or been announced.”
The implications of this are extremely exciting, especially for robo-advisors such as Betterment and others. These new algorithms could create new and better outcomes that may not even have to rely on exchange-traded funds (ETFs). That is just speculation, but if possible, it could allow for better performance even while avoiding fees charged by ETFs.
Human Input is Still a Must
Although big data has the potential to greatly improve investing decisions, that doesn’t mean it is a “set it and forget it” solution. These technologies are very powerful tools, but they are just that–tools. They must be carefully cultivated and calibrated to achieve the best outcome.
Anyone who has used an analytics or reporting tool understands this. They can provide an absolute wealth of information, but only if you set up the right reports and alerts. Nevertheless, big data will mean being able to understand market conditions in a way not previously possible.
“Portfolio managers exercise their judgment when selecting the data and analytics that we use in investing, and also when reviewing and approving each trade in every portfolio,” says Takashi Suwabe, a portfolio manager at GSAM. “This is to ensure that all portfolio positions make sense—that they are economically intuitive and appropriately sized given current market conditions. We do not have a computer in the corner simply shooting out trades with no human interaction.”
Suwabe goes on to say it’s not just about selecting an investment, but about finding a new way to select stocks. “Investment factors should be fundamentally-based and economically-motivated, and the data enables us to empirically test our investment hypotheses,” he says.” In other words, it’s not just about just trying to make better picks, but to improve the way in which you analyze investments. This is a more holistic approach that will become a reality with big data.
Investing for the Future–in the Future
There will be many changes in how companies do business in the 2020s. From increased automation to more business being conducted on eCommerce platforms, the way we have done things in the past won’t be the way forward.
And, indeed, big data provides a lot of new and exciting possibilities for investing. Although some investment managers do use some advanced metrics to analyze companies, they are nevertheless limited by human capability. The “datafication” of unstructured data combined with technologies such as machine learning and predictive analytics will greatly improve fund managers’ insights.
Human intervention is still needed, although that is probably preferable. Otherwise, you would have a computer placing large numbers of trades that you can only hope are not ill-advised. That being said, big data has the potential to revolutionize investing decisions and remove much of the speculation for those who prefer safer, more reliable growth.
Thus, we can improve investing outcomes by making better, more informed investments. Now, all we need is the data.