The Butterfly “Affect”: impact of development practices on cryptocurrency prices
File(s)s13688-020-00239-6.pdf (2.08 MB)
Published version
Author(s)
Type
Journal Article
Abstract
The network of developers in distributed ledgers and blockchains open source projects is essential to maintaining the platform: understanding the structure of their exchanges, analysing their activity and its quality (e.g. issues resolution times, politeness in comments) is important to determine how “healthy” and efficient a project is. The quality of a project affects the trust in the platform, and therefore the value of the digital tokens exchanged over it.
In this paper, we investigate whether developers’ emotions can effectively provide insights that can improve the prediction of the price of tokens. We consider developers’ comments and activity for two major blockchain projects, namely Ethereum and Bitcoin, extracted from Github. We measure sentiment and emotions (joy, love, anger, etc.) of the developers’ comments over time, and test the corresponding time series (i.e. the affect time series) for correlations and causality with the Bitcoin/Ethereum time series of prices. Our analysis shows the existence of a Granger-causality between the time series of developers’ emotions and Bitcoin/Ethereum price. Moreover, using an artificial recurrent neural network (LSTM), we can show that the Root Mean Square Error (RMSE)—associated with the prediction of the prices of cryptocurrencies—significantly decreases when including the affect time series.
In this paper, we investigate whether developers’ emotions can effectively provide insights that can improve the prediction of the price of tokens. We consider developers’ comments and activity for two major blockchain projects, namely Ethereum and Bitcoin, extracted from Github. We measure sentiment and emotions (joy, love, anger, etc.) of the developers’ comments over time, and test the corresponding time series (i.e. the affect time series) for correlations and causality with the Bitcoin/Ethereum time series of prices. Our analysis shows the existence of a Granger-causality between the time series of developers’ emotions and Bitcoin/Ethereum price. Moreover, using an artificial recurrent neural network (LSTM), we can show that the Root Mean Square Error (RMSE)—associated with the prediction of the prices of cryptocurrencies—significantly decreases when including the affect time series.
Date Issued
2020-12
Date Acceptance
2020-07-13
Citation
EPJ Data Science, 2020, 9 (1), pp.1-18
ISSN
2193-1127
Publisher
Springer
Start Page
1
End Page
18
Journal / Book Title
EPJ Data Science
Volume
9
Issue
1
Copyright Statement
© The Author(s) 2020. This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use,
sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original
author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other
third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by
statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a
copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original
author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other
third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by
statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a
copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
License URL
Identifier
https://epjdatascience.springeropen.com/articles/10.1140/epjds/s13688-020-00239-6
Publication Status
Published
Article Number
21
Date Publish Online
2020-07-23