Daniel Kahneman received the Nobel Prize in Economics for his research on prospect theory. His scholarship has shown how behavioral finance – and by extension sentiment evaluation – can improve our understanding of market behavior.
Sentiment evaluation applies algorithms to news articles, social media, and other data sources to evaluate how people feel in regards to the market, while behavioral economics identifies the cognitive biases that influence decision-making. Sentiment evaluation can assist make clear how these biases manifest themselves in financial markets. Of course, what people do is usually more revealing than what they are saying, so sentiment evaluation doesn’t all the time capture the complexity of human emotions in a field as charged as financial markets.
Still, it could help us interpret and predict market behavior. Here’s how.
Technical analysts are likely to measure sentiment tangentially, roughly estimating when a turning point will occur. However, their results are sometimes inconsistent because their methods are associative and should not discover the “cause” behind market results. Fundamental evaluation takes a more causal approach, but its feedback loop is usually longer than investors’ time frames and doesn’t all the time distinguish between value and a price trap.
The best investors intuitively understand that markets aren’t good at pricing in future outcomes. During the subprime crisis, for instance, the pricing of subprime securities suggested that the market valued 80% of the underlying loans at roughly zero. This resulted in a particularly favorable risk/reward ratio for those investors who knew what to search for. Last 12 months, too, market sentiment largely expected a recession this 12 months.
“The best trades are the ones that make you laugh on the set of CNBC.” — Jared Dillian
Jared Dillian is one among my favorite sentiment traders, and his perspective is essential. Although he believes in sentiment evaluation, he admits that it’s a difficult strategy for raising money. After all, a trade that inspires laughter doesn’t necessarily encourage confidence or investment capital. In addition, many doubt the scientific accuracy of sentiment evaluation and consider it to be comparable to astrology.
But by reorganizing market data and applying the principles of auction theory, we are able to use sentiment evaluation to categorise market behavior. James F Dalton has pioneered using the Market Profile technique developed by J. Peter Steidlmayer to discover the behavior of varied market participants. Specifically, Dalton’s technique observes the form of a day and other “market-generated information.” For example, if the market is falling on a specific day and only a limited variety of market participants are selling, or the sales are driven by long liquidations relatively than recent sellers, the form of the day might resemble the letter “b”. At the opposite end of the spectrum, the form could in the future resemble a letter “p” if speculation and short-covering activity drive buying. These behaviors indicate weaker forms of shopping for and selling and should indicate that the market might not be as strong or weak as price alone would suggest.
How can we know whether these shapes convey essential and actionable information? By using artificial intelligence (AI), we are able to test whether a day’s shape is resulting from a really random process. How? By modeling such a process and comparing it with the forms actually observed out there. If market movements are random, the distribution of shapes from a random process would match the actual distribution of shapes. But they do not try this.
Auction process: every day classification
The test shows with 99 percent certainty that these results don’t correspond to a really random process. If they aren’t arbitrary, they need to provide worthwhile information. In fact, the most important deviations from randomness occur when the shapes suggest that the market is simply too long and too short resulting from short covering or long liquidation. This supports the intuition that these behaviors are each unique and potentially actionable from an investment perspective.
In “Market Profile with Convolutional Neural Networks: Learning the Structure of Price Activities,“Chern-Bin Ju, Min-Chih Hung and An-Pin Chen show that similar image recognition techniques will be used to discover market patterns that may function the premise for commodity producers’ hedging strategies. Such research may lead to a deeper understanding of the market’s pricing process and help quantify investor sentiment. Investors are likely to give attention to price alone and momentum strategies are widely followed. Such stores may grow to be too crowded at times, which can lead to reversals. This isn’t random behavior, and now now we have a solution to objectively measure this behavior.
This research provides a heuristic-based technique for causality testing. Markus Schuller and Andreas Haberl have presented the groundbreaking arguments for causality in “Causality Techniques in Investment Management: Five Key Findings”. They note that financial markets are “complex, dynamic and forward-looking” and driven by “market participants with incomplete information and limited rationality.” The opportunity to objectively observe and measure the behavior of those market participants is “both attractive and potentially very lucrative.”
Here’s why sentiment evaluation can assist uncover alpha opportunities and why it’s value adding to our investment toolkits.
If you enjoyed this post, remember to subscribe.
Photo credit: ©Getty Images / vladystock
