What kind of role does sentiment play in the markets? Many believe that the psychological and emotional response to investing and trading decisions are where there is a significant edge to be had.
And Thomson Reuters agrees. The media firm has expanded its data sentiment offerings to track the top 100 cryptocurrencies. In association with MarketPsych Indices, the TRMI package uses machine learning and natural language processing to measure emotional and topical items from across news and social media sites that have the potential to drive market participant behaviour in cryptocurrency markets.
Following the successful launch of bitcoin sentiment data in March 2018, Thomson Reuters new offering will monitor more than 2,000 global news and 800 social media sites in real-time, yielding 43 themes and sentiments on the top 100 cryptocurrencies.
TRMI is accompanied by visualization tools and a suite of quantitative research results developed by MarketPsych, including regression and cross-sectional rotation models, to help traders identify influential themes and more rapidly develop actionable strategies. Historical data dates back to 2009 and will help to identify the predictive value of the data.
A New Edge
Eric Fischkin Proposition Director, Machine Readable News at Thomson Reuters and Dr Richard Peterson CEO, MarketPsych are two of the key figures behind the new sentiment indicator. They feel this new tool will only benefit investors by offering insight that was previously unavailable.
“The sentiment indicators measure the emotions expressed and the themes discussed around specific equities, commodities, and currencies (among other entities). They convert these themes and emotions into time series that can be used in trading models,” said Fischkin and Peterson.
“This allows investors to compare the emotions of investors about specific currencies. For example, which ones have a higher frequency of chatter about “FOMO” or being a “Scam”. As a result, they can make better investment decisions.”
While markets are saturated with data on technical and fundamental analysis. Using sentiment is a new way of looking at the ratio of supply and demand. And Fischkin and Peterson believe this might be a new way to gain an edge.
“We see that information flow often moves markets. Yet information flow occurs on many platforms and is often difficult to detect. If we can quantify the salient aspects of that information flow in millions of articles, then new possibilities of using machines to assist human decision making open up,” said Fischkin and Peterson.
While the indicator is still new there are a few ways Fischkin and Peterson suggest it be used to maximize its power.
“There are many techniques, but we recommend specific types of visualizations as well as machine learning techniques to identify the best predictive power in the data. Our clients who use the data for equities identified a simple strategy for identifying tops that also appears useful for cryptocurrencies,” said Fischkin and Peterson.
Cryptocurrencies Becoming Mainstream
The suite of tools to help better understand Cryptocurrencies markets comes at an important time. According to a Thomson Reuters survey released in April 2018, approximately 20% of financial firms indicated they are considering trading cryptocurrency over the next 3-12 months. However, providing trading insight to the cryptocurrency market is unique, as online communications and information flow are significant drivers of cryptocurrency values, in comparison to traditional financial services assets.
Sentiment analysis of this market therefore often requires understanding the top cryptocurrencies at any given time, where individuals get their information, which digital platforms are used for communication, and how specific language or terms used may signify future trends (for example: FOMO, HODL, etc.).
