This week saw the announcement that Tesla will be including self-driving capability into all their new cars. Few will argue that the growing burden of traffic, vehicular fatalities, the resulting impacts on our environment and the large time suck that the commute can cause are global challenges worth solving. However, as we have also seen over the last year with a variety of high profile autonomous driving accidents; there is capability in the human machine that continues to surpass the current generation of machine computing capability.
Whether its driving, or trading the global financial markets there is extensive value to be gained by utilizing machines. But we cannot lose sight of the value of the human element of investing. As Bruce Fador commented in Technology Can’t Fully Replace Human Judgement and Intuition, "In many corners of the market, lucrative buy and sell opportunities reveal themselves with things that aren’t technical indicators, and aren’t easily captured and fed into models".
Given the contrast of a fully automated world and a fully human world it is crucial that capital market participants not become luddites. Embracing the progress in machine learning, quantitative & predictive models, and advances yet to come while also capturing the human element of investing and trading is crucial to finding and delivering alpha.
So, how does a firm embrace technology while retaining the benefit of human intuition? One approach is to apply natural-language processing, combined with elements of machine learning to perform sentiment analysis. This process captures the human benefit of gauging tone, elements of meaning and comprehending when those things change. In the past a firm could only rely on the amount of textual data that individuals could process or consume for meaning and trends, for example how did the CEO sound, what descriptors did they use as they spoke, during the quarterly earnings call - how did they sound during the previous earning calls. Now, by employing machine learning and processing techniques that implement the very human capability of processing sentiment, a firm can process and integrate into their entire investment framework a volume of data that has been beyond the reach of all but the largest staffed research departments. This quantification of textual big-data as sentiment values not only provides a view into much larger sets of data but also provides a means by which to measure changes over time of sentiment from specific sources, companies, and markets.
From quantitative hedge fund to traditional investment management firms to individual market participants the keys to success in the market are to employ technology, both human and machine, to measure the entire market environment (stock price, volume, earnings, news, sentiment and more), identify changes in that market environment, and act ahead of the impact those changes will have on the capital markets. By embracing technology quants, traders, and investment firms extend their capabilities to consume and integrate, effectively, much larger amounts of data into their entire process - empowering them to capitalize on these keys to success and provide differentiating alpha.