How is big data impacting the world of finance?
It’s definitely big — and it’s bringing finance into the 21st century through access to more real-time and precise ways to understand an increasingly complex world.
Advanced analytics, superior processing, trained human judgement and innovation are playing key roles in developing a consistent and investable competitive edge, says Osman Ali, portfolio manager of quantitative investment strategies at Goldman Sachs Asset Management.
When we think of alternative, unstructured, and sometimes unconventional sources of data, Ali says, one may think they are hard to come by and hard to process based on size.
“But if you have the right technology and tools looking at it, you can find some interesting insights,” he says.
Never before have investors and corporations faced the growing deluge of data that is produced globally — every second. And, in the next few years, the amount of data produced will get exponentially larger.
Given the huge and evolving data landscape, Ali asks whether or not big data and technology can be used to better calculate inflation.
The answer is yes.
The Billion Prices Project (BPP), an initiative designed and tested at the MIT Sloan School of Management, is using big data to better estimate inflation by gathering prices from hundreds of online retailers around the world on a daily basis. That’s a significant leap from the 1950s, when collection was done through in-person notation.
“BPP is a very intuitive and efficient way of building a real-time inflation indicator,” he says. “If you are an investor, at the very least, it gives you a forward-looking indication of what that Consumer Price Index number might look like.”
THE VIEW FROM SPACE
Another example is luminosity, or an analysis technique which uses satellite images to examine light emissions coming out of urban areas, showing how they change over time. He says urban sprawl and population growth facts can be collected and correlated with statistics already published.
“Imagine the power of having this information in real time, well before the market” he says.
At the stock selection level, machine reading can be being used to process material that is unstructured. Ninety percent of the world’s data is non-numeric and unstructured. Audio recordings, news articles and research reports, in various languages, are now able to be analyzed.
Ali left the audience thinking about new machine learning technologies like natural language processing, which allows computers to collect insights from sources such as real-time news, earnings calls, social media, videos and images more quickly than ever.
Quantifying factors, like the sentiment of the market view and how people are thinking, provides fundamental analysis for understanding what people are saying. And using sentiment analysis, Ali says, gives investors a good picture, on net, of what people think about companies.
Such advanced analytics and superior processing technology are the keys to extracting value and actionable perceptions from this flood of information, Ali concluded. As the rate of data production increases, human judgement will become an increasingly essential tool in harnessing the power of big data.