I am happy to announce the release of a working paper which I have written with my colleagues Teo Geldner and Joscha Beckmann entitled:
"The relevance of sentiment and media attention for bitcoin holdings across entities."
You can read the full paper here:
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4436838
Too long didn’t read:
- small investors react strongly to public information and media narratives
- reaction of large-scale investors much weaker
- large-scale investors appear to rely more on non-public information, and they may have different investment objectives than small-scale investors
Long summary:
Our study investigates the relationship between sentiment, media attention, and bitcoin holdings among different entities using data from Glassnode for entity information and Thomson Reuters MarketPsych for sentiment data.
We differentiate between small and large entities and analyze how they react to incoming public information. We use linear break point regressions to account for potential shifts in the composition of the type of holders and/or changes in how bitcoin is treated as an asset over time.
Our main findings indicate that price dynamics and sentiment changes have varying effects on bitcoin holdings depending on the size of the entity's holdings. Small (retail) investors respond strongly to public information and media narratives, with their decisions strongly influenced by sentiment and media attention, while large-scale investors respond more weakly to such information.
Additionally, small-scale investors tend to increase their positions when bitcoin price increases, sentiment is positive, and buzz is high, while larger investors tend to react in opposing directions. These results are consistent with our expectations, as large-scale investors may have access to different non-public information for their investment decisions, and they may have different investment objectives compared to retail investors.
The literature on investor behavior in the Bitcoin space is still relatively
scarce, and the use of on-chain data in the economics field is also highly
limited. We hope our study can contribute to the literature by shedding more light on investor behavior in the bitcoin space and drawing attention to the blockchain as a valuable data source for academic research.
Finally, I would like to give a huge shout-out to Glassnode for compiling such a valuable data source.