在過去，經濟學家認為情緒的改變無法大幅度的對股市價格產生影響，但隨
著不斷的發展，越來越多人開始對市場情緒研究，試圖檢查其是否以任何方式
造成了目前仍無法解釋的市場價格變化。
本論文針對全世界加密貨幣市場中交易量最大的「比特幣」(Bitcoin) 作出研
究，資料集為2020年到2022年的歷史交易資料。這一時期的特點是比特幣價格
發生了顯著變化，投資者情緒也存在巨大變化。我們模擬了D’Agostino (2021)
開發出來的模型，並參考了近年來針對加密貨幣提出來的情緒指標「Crypto
Fear and Greed Index」分析比特幣價格的變化與市場情緒有無相關。本篇研究
展示當我們參考了Crypto Fear and Greed Index時，可改良比特幣漲跌的預測，
使我們的模型相較D’Agostino之原始模型更加準確。

In the past, economists believed that changes in market sentiment could not
make significant impacts on stock prices. However, as time goes by, more and
more research were conducted on market sentiment, checking whether it in any
way creates those price changes that so far cannot be explained.
In this thesis, we investigate Bitcoins (Nakamoto, 2008), the most traded cryptocurrency in the world, on its pricing from 2020 to 2022. This period witnessed
significant fluctuations in both Bitcoin prices and market sentiment. We simulate
the model developed by D’Agostino (2021), and augment it with Crypto Fear and
Greed Index, which is a recently proposed sentiment indicator for crypocurrency
market, to investigate whether there is any relationship between Bitcoin price
changes and market sentiment.
Our results demonstrate that taking Crypto Fear and Greed Index into account
can steadily improve Bitcoin’s price-up and price-down prediction, thus producing
a more accurate model than D’Agostino’s original model.

[1] Muhammad Amjad and Devavrat Shah. Trading Bitcoin and Online Time
Series Prediction. In Proceedings of NIPS 2016 Time Series Workshop, pages
1–15, 2016.
[2] David Bourghelle, Fredj Jawadi, and Philippe Rozin. Do Collective Emotions Drive Bitcoin Volatility? A Triple Regime-Switching Vector Approach.
Journal of Economic Behavior and Organization, 196:294–306, 2022.
[3] Jaroslav Bukovina and Mat´uˇs Marticek. Sentiment and Bitcoin Volatility.
MENDELU Working Papers in Business and Economics 2016-58, Mendel
University of Brno, 2016.
[4] Conghui Chen, Lanlan Liu, and Ningru Zhao. Fear Sentiment, Uncertainty,
and Bitcoin Price Dynamics: The Case of COVID-19. Emerging Markets
Finance and Trade, 56(10):2298–2309, 2020.
[5] Tianqi Chen and Carlos Guestrin. XGBoost: A Scalable Tree Boosting System. In Proceedings of ACM SIGKDD International Conference on Knowledge
Discovery and Data Mining, pages 785–794, 2016.
[6] Zheshi Chen, Chunhong Li, and Wenjun Sun. Bitcoin Price Prediction using
Machine Learning: An Approach to Sample Dimension Engineering. Journal
of Computational and Applied Mathematics, 365:Article 112395, 2020.
[7] Giuseppe D’Agostino. Study and Development of Machine Learning-Based
Cryptocurrency Trading Systems. Master’s thesis, Politecnico di Torino, 2021.
http://webthesis.biblio.polito.it/id/eprint/19177.
[8] Alex Greaves and Benjamin Au. Using the Bitcoin Transaction Graph to Predict the Price of Bitcoin, 2015. Available at:
https://snap.stanford.edu/class/cs224w-2015/projects 2015/.
[9] Vytautas Karalevicius, Niels Degrande, and Jochen De Weerdt. Using Sentiment Analysis to Predict Interday Bitcoin Price Movements. Journal of Risk
Finance, 19(1):56–75, 2018.
[10] Sean McNally, Jason Roche, and Simon Caton. Predicting the Price of Bitcoin
using Machine Learning. In Euromicro Conference on Parallel, Distributed
and Network-Based Processing, pages 339–343, 2018.
[11] Satoshi Nakamoto. Bitcoin: A Peer-to-Peer Electronic Cash System, 2008.
https://bitcoin.org/bitcoin.pdf.