Accounting for Human Error: Strategizing Sentiment for Cryptocurrency

Tyler Sack
September 9, 2019

Behavioural finance is an interdisciplinary study in economics that uses psychology based theories to explain how and why people make financial decisions. Specifically, it attempts to explain why people act irrationally when it comes to money.

The field addresses a gap in conventional economic and financial theory that fails to capture the extent that human biases, emotions, and cognitive reasoning influence or prevent us from making the most logical decisions when it comes to money. Early research in the field was concerned with explaining the anomalies or paradoxes that could not be adequately explained by conventional models.

Attributing errors and inconsistencies in prices to human behaviour has led to the development of theories explaining confirmation bias, herd instinct behaviour, gambling fallacy, anchoring, loss aversion, and mental accounting. All of which explain why people make decisions that are contrary or riskier than others.

There has been a shift in financial theory and strategy to look outside of traditional asset pricing models to consider behavioural-based analysis. The benefits of having a blended approach is about challenging assumptions, checking biases, and seeking out more information that tests the objectivity of findings. Being aware of the behaviours that can cause errors should lead people to at least create systems to check their own set of conceptual baggage to reduce human error and manage assets rationally.

When considering cryptocurrency, however, there remains a lot of speculation and uncertainty on the topic of strategy. While there have been successful individuals and even firms who made incredible returns on their investments, we are unsure of the underlying conditions that determine future prices of cryptocurrency. Outside of the supply/demand, there is no set of key drivers that indicate how cryptocurrency, or more accurately the people who own cryptocurrency, will react to market or government developments. Without any intrinsic value or method of determining asset price, cryptocurrency values may be best measured by the sentiments of those who are directly involved in trading.

Without relying on fundamentals, investors may have to gather as much information from social media, news, and other market trends to develop context for understanding why the market reacted the way it did in that particular time. Basically, combining the technical analysis of historical prices with the context of sentiment and behaviour errors may be the most reliable strategy for investing in cryptocurrency.

At Orenda, our goal is to understand this idea. We are analyzing several datasets on cryptocurrency to test whether sentiment is an indicator of value. It isn’t that we believe sentiment is a predicator of price, but that we can infer human behaviour and irrational decision making from social media sentiment.