# Segmenting Bitcoin Transactions for Price Movement Prediction

^{1}

^{2}

^{3}

^{4}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Neither Fish nor Fowl: A Review of Challenges in Current Predictions of Bitcoin Price Movements

## 3. Not All Blocks Are Created Equal: Investor Segmentation

## 4. Data Analysis and Results

#### 4.1. Data

#### 4.2. Transaction Segmentation

#### 4.3. Price Movement Prediction

- (1)
- ${\overrightarrow{\mathbf{m}}}_{T-1}$, a vector representing the direction of the historical Bitcoin price movement from time $T-p$ to $T-1$. This vector of zeros and ones is denoted by:$${\overrightarrow{m}}_{T-1}=({m}_{T-1},{m}_{T-2},\cdots ,{m}_{T-p})$$
- (2)
- $\Delta {\overrightarrow{\mathbf{N}}}_{T-1}$, a vector representing the change in transaction volume of all transaction classes during each time period from time $T-p$ to $T-1$. This vector is denoted by:$$\begin{array}{cc}\hfill \Delta {\overrightarrow{\mathbf{N}}}_{T-1}& ={(\Delta {n}_{i,t})}_{(1\le i\le 15,T-p\le t\le T-1)}\hfill \end{array}$$$$\begin{array}{cc}\hfill \Delta {\overrightarrow{\mathbf{N}}}_{T-1}& =\left(\begin{array}{c}\Delta {n}_{1,T-1},\cdots ,\Delta {n}_{15,T-1}\\ \cdots \\ \Delta {n}_{1,T-(p-1)},\cdots ,\Delta {n}_{15,T-(p-1)}\\ \Delta {n}_{1,T-p},\cdots ,\Delta {n}_{15,T-p}\end{array}\right)\hfill \end{array}$$
- (3)
- A variable ${F}_{T}$, representing a fixed effect measured at time T. We consider two types of time-specific fixed effects: the day/night fixed effects and the month fixed effects. We tested our results with various values of the look-back period (p). For consistency and simplicity, and to save space, we present the subsequent results only for the case $p=10$.

#### 4.4. Results

#### 4.5. The Relationship between Price Movement and Transaction Volume in the Bitcoin Market

## 5. Interpretation and Discussion

#### 5.1. Market Participants

#### 5.1.1. Individual Investors

#### 5.1.2. Algorithmic Bitcoin Traders (Micro-Traders)

#### 5.1.3. Larger-Volume Traders (Institutions and Whales)

#### 5.2. Using Prediction of Price Direction Movement as an Investment Guide

## 6. Conclusions and Implications

## Author Contributions

## Funding

## Data Availability Statement

## Conflicts of Interest

## Notes

1 | In fairness, it should be noted that many “fiat currencies” issued by governments may also sometimes fail to satisfy one or more of these functions. |

2 | However, when there is a lack of trust in the local fiat currency, which sometimes occurs in some developing or emerging economies, or when investment in foreign currency is forbidden, storing value in Bitcoin may be attractive despite its high volatility. |

3 | Cheah and Fry (2015) showed that Bitcoin exhibits speculative bubbles with a fundamental price of zero. |

4 | We consider T to be the time when a block is mined on the Bitcoin blockchain, which is approximately every 10 min. |

5 | Studies have shown that in many contexts neural network models outperform traditional statistical models for prediction, and that logistic regression is among the best (and easiest to explain) of the of the traditional statistical classification methods (cf., West et al. 1997; Brockett et al. 1994, 2006). |

6 | |

7 | Our dataset uses Bitcoin information from 2011 to 2017; however, much has evolved in the Bitcoin market since then (e.g., the emergence of Exchange Traded Funds for Bitcoin, El Salvador recognizing Bitcoin as legal tender for transactions, etc.). Nevertheless, there is nothing in these changes that would cast doubt on the conclusion of this paper that there is an accuracy benefit to using segmented transaction data in conjunction with Bitcoin price data to better predict the direction of Bitcoin price movement. Additionally, while we recognize that using data from 2011 to 2017 may raise questions about topicality, we note that using a newer dataset can pose additional challenges because a large number of transactions have been happening off-chain since 2018 (https://www.chainalysis.com/blog/fake-trade-volume-cryptocurrency-exchanges/ (accessed on 1 March 2020)). Many of the crypto exchanges provided a channel for investors to trade without registering the transaction on the BTC blockchain. |

## References

- Aalborg, Halvor Aarhus, Peter Molnár, and Jon Erik de Vries. 2019. What can explain the price, volatility and trading volume of bitcoin? Finance Research Letters 29: 255–65. [Google Scholar] [CrossRef]
- Abraham, Jethin, Daniel Higdon, John Nelson, and Juan Ibarra. 2018. Cryptocurrency price prediction using tweet volumes and sentiment analysis. SMU Data Science Review 1. No. 3. Available online: https://scholar.smu.edu/datasciencereview/vol1/iss3/1 (accessed on 10 March 2024).
- Adrian, Tobias, and Tommaso Mancini-Griffoli. 2021. The rise of digital money. Annual Review of Financial Economics 13: 57–77. [Google Scholar] [CrossRef]
- Akcora, Cuneyt G., Asim Kumer Dey, Yulia R. Gel, and Murat Kantarcioglu. 2018. Forecasting bitcoin price with graph chainlets. In Advances in Knowledge Discovery and Data Mining: 22nd Pacific-Asia Conference, PAKDD 2018, Proceedings, Part III 22, Melbourne, VIC, Australia, June 3–6. Berlin and Heidelberg: Springer, pp. 765–76. [Google Scholar]
- Almeida, José, and Tiago Cruz Gonçalves. 2023. Portfolio diversification, hedge and safe-haven properties in cryptocurrency investments and financial economics: A systematic literature review. Journal of Risk and Financial Management 16: 3. [Google Scholar] [CrossRef]
- Alvarez, Fernando, David Argente, and Diana Van Patten. 2023. Are cryptocurrencies currencies? bitcoin as legal tender in el salvador. Science 382: eadd2844. [Google Scholar] [CrossRef]
- Amjad, Muhammad J., and Devavrat Shah. 2016. Nips time series workshop. In Trading Bitcoin and Online Time Series Prediction, Barcelona, Spain, December 9. PMLR. pp. 1–15. Available online: https://proceedings.mlr.press/v55/amjad16.html (accessed on 1 March 2024).
- Arguelles, Dennys Christian Mallqui. 2018. Predicting the Direction, Maximum, Minimum and Closing Price of Daily/Intra-Daily Bitcoin Exchange Rate Using Batch and Online Machine Learning Techniques. Master’s thesis, Universidade Federal de São Carlos, São Carlos, Brazil. Available online: https://repositorio.ufscar.br/handle/ufscar/10589 (accessed on 1 March 2020).
- Arnosti, Nick, and Matthew Weinberg. 2022. Bitcoin: A natural oligopoly. Management Science 68: 4755–71. [Google Scholar] [CrossRef]
- Atsalakis, George S., Ioanna G. Atsalaki, Fotios Pasiouras, and Constantin Zopounidis. 2019. Bitcoin price forecasting with neuro-fuzzy techniques. European Journal of Operational Research 276: 770–80. [Google Scholar] [CrossRef]
- Bakry, Walid, Audil Rashid, Somar Al-Mohamad, and Nasser El-Kanj. 2021. Bitcoin and portfolio diversification: A portfolio optimization approach. Journal of Risk and Financial Management 14: 282. [Google Scholar] [CrossRef]
- Bariviera, Aurelio F. 2017. The inefficiency of bitcoin revisited: A dynamic approach. Economics Letters 161: 1–4. [Google Scholar] [CrossRef]
- Bartolucci, Silvia, Giuseppe Destefanis, Marco Ortu, Nicola Uras, Michele Marchesi, and Roberto Tonelli. 2020. The butterfly “affect”: Impact of development practices on cryptocurrency prices. EPJ Data Science 9: 255–65. [Google Scholar] [CrossRef]
- Battey, Heather, David Cox, and Michelle Jackson. 2019. On the linear in probability model for binary data. Royal Society Open Science 6: 190067. [Google Scholar] [CrossRef]
- Blau, Benjamin M. 2017. Price dynamics and speculative trading in bitcoin. Research in International Business and Finance 41: 493–99. [Google Scholar] [CrossRef]
- Brockett, Patrick L., Linda L. Golden, Jaeho Jang, and Chuanhou Yang. 2006. A comparison of neural network, statistical methods, and variable choice for life insurers’ financial distress prediction. Journal of Risk and Insurance 73: 397–419. [Google Scholar] [CrossRef]
- Brockett, Patrick L., William W. Cooper, Linda L. Golden, and Utai Pitaktong. 1994. A neural network method for obtaining an early warning of insurer insolvency. The Journal of Risk and Insurance 61: 402–24. [Google Scholar] [CrossRef]
- Chan, Louis K. C., and Josef Lakonishok. 1995. The behavior of stock prices around institutional trades. The Journal of Finance 50: 1147–74. [Google Scholar] [CrossRef]
- Cheah, Eng-Tuck, and John Fry. 2015. Speculative bubbles in bitcoin markets? an empirical investigation into the fundamental value of bitcoin. Economics Letters 130: 32–36. [Google Scholar] [CrossRef]
- Chen, Junwei. 2023. Analysis of bitcoin price prediction using machine learning. Journal of Risk and Financial Management 16: 51. [Google Scholar] [CrossRef]
- Dimpfl, Thomas, and Stefania Odelli. 2020. Bitcoin price risk—A durations perspective. Journal of Risk and Financial Management 13: 157. [Google Scholar] [CrossRef]
- Dutta, Aniruddha, Saket Kumar, and Meheli Basu. 2020. A gated recurrent unit approach to bitcoin price prediction. Journal of Risk and Financial Management 13: 23. [Google Scholar] [CrossRef]
- Dyhrberg, Anne H., Sean Foley, and Jiri Svec. 1995. How investible is bitcoin? analyzing the liquidity and transaction costs of bitcoin markets. Economics Letters 171: 140–43. [Google Scholar] [CrossRef]
- Edelen, Roger M., Ozgur S. Ince, and Gregory B. Kadlec. 2016. Institutional investors and stock return anomalies. Journal of Financial Economics 119: 472–88. [Google Scholar] [CrossRef]
- Fischer, Thomas Günter, Christopher Krauss, and Alexander Deinert. 2019. Statistical arbitrage in cryptocurrency markets. Journal of Risk and Financial Management 12: 31. [Google Scholar] [CrossRef]
- Gabaix, Xavier, Parameswaran Gopikrishnan, Vasiliki Plerou, and Harry Eugene Stanley. 2006. Institutional investors and stock market volatility. Quarterly Journal of Economics 121: 461–504. [Google Scholar] [CrossRef]
- Gandal, Neil, J. T. Hamrick, Tyler Moore, and Tali Oberman. 2018. Price manipulation in the bitcoin ecosystem. Journal of Monetary Economics 95: 86–96. [Google Scholar] [CrossRef]
- Geuder, Julian, Harald Kinateder, and Niklas F. Wagner. 2019. Cryptocurrencies as financial bubbles: The case of bitcoin. Finance Research Letters 31. [Google Scholar] [CrossRef]
- Griffin, John M., and Amin Shams. 2020. Is bitcoin really untethered? Journal of Finance 75: 1913–64. [Google Scholar] [CrossRef]
- Gronwald, Marc. 2019. Is bitcoin a commodity? on price jumps, demand shocks, and certainty of supply. Journal of International Money and Finance 97: 86–92. [Google Scholar] [CrossRef]
- Gsell, Markus. 2008. Assessing the Impact of Algorithmic Trading on Markets: A Simulation Approach (2008). 16th European Conference on Information Systems (ECIS) 2008 Proceedings, 225. Galway, Ireland; Available online: https://aisel.aisnet.org/ecis2008/225 (accessed on 1 March 2024).
- Holden, Richard, and Anup Malani. 2022. An examination of velocity and initial coin offerings. Management Science 68: 9026–41. [Google Scholar] [CrossRef]
- Jang, Jeewon, and Jangkoo Kang. 2019. Probability of price crashes, rational speculative bubbles, and the cross-section of stock returns. Journal of Financial Economics 132: 222–47. [Google Scholar] [CrossRef]
- Ji, Li Jun, Zhiyong Zhang, and Tieyuan Guo. 2008. To buy or to sell: Cultural differences in stock market decisions based on price trends. Journal of Behavioral Decision Making 21: 399–413. [Google Scholar] [CrossRef]
- Kinderis, Marius, Marija Bezbradica, and Martin Crane. 2018. Bitcoin currency fluctuation. In Proceedings of the 3rd International Conference on Complexity, Future Information Systems and Risk—Volume 1: COMPLEXIS. Setúbal: INSTICC, SciTePress, pp. 31–41. [Google Scholar]
- Kjærland, Frode, Aras Khazal, Erlend A. Krogstad, Frans B. G. Nordstrøm, and Are Oust. 2018. An analysis of bitcoin’s price dynamics. Journal of Risk and Financial Management 11: 63. [Google Scholar] [CrossRef]
- Koutmos, Dimitrios. 2018. Bitcoin returns and transaction activity. Economics Letters 167: 81–85. [Google Scholar] [CrossRef]
- Kurbucz, Marcell Tamás. 2019. Predicting the price of bitcoin by the most frequent edges of its transaction network. Economics Letters 184: 108655. [Google Scholar] [CrossRef]
- Kyriazis, Nikolaos A. 2019. A survey on efficiency and profitable trading opportunities in cryptocurrency markets. Journal of Risk and Financial Management 12: 67. [Google Scholar] [CrossRef]
- Li, Jing-Ping, Bushra Naqvi, Syed Kumail Abbas Rizvi, and Hsu-Ling Chang. 2021. Bitcoin: The biggest financial innovation of fourth industrial revolution and a portfolio’s efficiency booster. Technological Forecasting and Social Change 162: 120383. [Google Scholar] [CrossRef]
- Lyandres, Evgeny, Berardino Palazzo, and Daniel Rabetti. 2022. Initial coin offering (ico) success and post-ico performance. Management Science 68: 8658–79. [Google Scholar] [CrossRef]
- Madan, Isaac, Shaurya Saluja, and Aojia Zhao. 2014. Automated Bitcoin Trading via Machine Learning Algorithms. Available online: https://api.semanticscholar.org/CorpusID:14217274 (accessed on 1 March 2020).
- Mai, Feng, Zhe Shan, Qing Bai, Xin (Shane) Wang, and Roger Chiang. 2018. How does social media impact bitcoin value? a test of the silent majority hypothesis. Journal of Management Information Systems 35: 19–52. [Google Scholar] [CrossRef]
- McNally, Sean, Jason Roche, and Simon Caton. 2018. Predicting the price of bitcoin using machine learning. Paper presented at 2018 26th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP), Cambridge, UK, March 21–23; pp. 339–43. [Google Scholar]
- Meese, Richard A., and Kenneth Rogoff. 1983. Empirical exchange rate models of the seventies: Do they fit out of sample? Journal of International Economics 14: 3–24. [Google Scholar] [CrossRef]
- Miller, Dante, and Jong-Min Kim. 2021. Univariate and multivariate machine learning forecasting models on the price returns of cryptocurrencies. Journal of Risk and Financial Management 14: 486. [Google Scholar] [CrossRef]
- Moosa, Imad, and Kelly Burns. 2014. The unbeatable random walk in exchange rate forecasting: Reality or myth? Journal of Macroeconomics 40: 69–81. [Google Scholar] [CrossRef]
- Nadarajah, Saralees, and Jeffrey Chu. 2017. On the inefficiency of bitcoin. Economics Letters 150: 6–9. [Google Scholar] [CrossRef]
- Niederhoffer, Victor, and Matthew Fontaine Maury Osborne. 1966. Market making and reversal on the stock exchange. Journal of the American Statistical Association 61: 897–916. [Google Scholar] [CrossRef]
- O’Connell, Paul G. J., and Melvyn Teo. 2009. Institutional investors, past performance, and dynamic loss aversion. Journal of Financial and Quantitative Analysis 44: 155–88. [Google Scholar] [CrossRef]
- Popper, Nathaniel. 2018. As Bitcoin Bubble Loses Air, Frauds and Flaws Rise to Surface. The New York Times. February 5. Available online: https://www.nytimes.com/2018/02/05/technology/virtual-currency-regulation.html (accessed on 1 March 2020).
- Sattarov, Otabek, Heung Seok Jeon, Ryumduck Oh, and Jun Dong Lee. 2020. Forecasting bitcoin price fluctuation by twitter sentiment analysis. Paper presented at 2020 International Conference on Information Science and Communications Technologies (ICISCT), Tashkent, Uzbekistan, November 4–6. [Google Scholar]
- Scharnowski, Stefan. 2021. Understanding bitcoin liquidity. Journal of Financial and Quantitative Analysis 38: 101477. [Google Scholar] [CrossRef]
- SEC, US Securities Exchange Commission. 2014. Investor Alert: Bitcoin and Other Virtual Currency-Related Investments. Available online: https://www.sec.gov/oiea/investor-alerts-bulletins/investoralertsia_bitcoin (accessed on 1 March 2020).
- Sin, Edwin, and Lipo Wang. 2017. Bitcoin price prediction using ensembles of neural networks. Paper presented at 2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD), Guilin, China, July 29–31; pp. 666–671. [Google Scholar]
- Su, Chi-Wei, Meng Qin, Ran Tao, and Muhammad Umar. 2020. Financial implications of fourth industrial revolution: Can bitcoin improve prospects of energy investment? Technological Forecasting and Social Change 158: 120178. [Google Scholar] [CrossRef]
- Weaver, Nicholas. 2018. Risks of cryptocurrencies. Communications of the ACM 61: 20–24. [Google Scholar] [CrossRef]
- West, Patricia M., Patrick L. Brockett, and Linda L. Golden. 1997. A comparative analysis of neural networks and statistical methods for predicting consumer choice. Marketing Science 16: 370–91. [Google Scholar] [CrossRef]

Ref. | Data | Prediction Method | Prediction Period | Best Accuracy |
---|---|---|---|---|

Madan et al. (2014) | Historical Price | Generalized Linear Model | Next 10 min | 0.57 |

Sattarov et al. (2020) | Twitter + Historical Price | Sentiment Analysis | Next 30 min | 0.62 |

Sin and Wang (2017) | Historical Price | Artificial Neural Network | Next Day | 0.60 |

McNally et al. (2018) | Historical Price | Long Short-Term Memory | Next Day | 0.52 |

Kinderis et al. (2018) | Social Media (Twitter) + Historical Price | Linear Discriminant Analysis | Next Day | 0.67 |

Atsalakis et al. (2019) | Historical Price | Artificial Neural Network PATSOS Neuro-Fuzzy Controller Forecasting | Next Day | 0.67 |

Arguelles (2018) | Historical Price | Support Vector Machine | Next Day | 0.62 |

Kurbucz (2019) | Transaction Network | Single Hidden-Layer Feedforward Neural Networks | Next Day | 0.60 |

Transaction Amount Segment | Total # of Transactions | Mean # of Transactions per Block |
---|---|---|

(0, 0.1] BTCs | 145,731,106 | 448 |

(0.1, 0.5] BTCs | 38,978,755 | 120 |

(0.5, 1] BTCs | 12,944,521 | 40 |

(1, 5] BTCs | 18,690,349 | 57 |

(5, 10] BTCs | 5,199,054 | 16 |

(10, 25] BTCs | 5,071,494 | 16 |

(25, 50] BTCs | 2,677,391 | 8 |

(50, 100] BTCs | 1,592,056 | 5 |

(100, 200] BTCs | 887,452 | 3 |

(200, 500] BTCs | 758,835 | 2 |

(500, 1000] BTCs | 226,882 | 1 |

(1000, 5000] BTCs | 150,458 | 0.5 |

(5000, 10,000] BTCs | 37,608 | 0.1 |

(10,000, 50,000] BTCs | 18,042 | 0.05 |

(50,000, ∞] BTCs | 529 | 0.002 |

Direction of Bitcoin Price Movement Prediction | ||||||

Using Transaction Volume Changes | ||||||

Model | Hist. Price | Trans. Volume | Month F.E. | Day/Night F.E. | Accuracy | F-1 Score |

LSTM | Yes | Yes | No | Yes | 0.631 | 0.773 |

LSTM | Yes | Yes | Yes | Yes | 0.631 | 0.768 |

Logistic Regression | No | Yes | No | No | 0.599 | 0.663 |

Logistic Regression | Yes | Yes | Yes | Yes | 0.596 | 0.652 |

Logistic Regression | Yes | No | Yes | Yes | 0.532 | 0.585 |

Logistic Regression | Yes | No | No | No | 0.496 | 0.497 |

Direction of Bitcoin Price Movement Prediction | ||||||

Using Transaction Proportion Distribution Changes | ||||||

Model | Hist. Price | Trans. Volume | Month F.E. | Day/Night F.E. | Accuracy | F-1 Score |

LSTM | Yes | Yes | Yes | Yes | 0.636 | 0.778 |

LSTM | Yes | Yes | No | Yes | 0.635 | 0.777 |

Logistic Regression | Yes | Yes | Yes | Yes | 0.538 | 0.586 |

Logistic Regression | Yes | Yes | No | Yes | 0.535 | 0.592 |

Segment | t − 1 | t − 2 | t − 3 | t − 4 | t − 5 | t − 6 |
---|---|---|---|---|---|---|

(0, 0.1) | $4.52\times {10}^{-5}$ *** | $4.14\times {10}^{-5}$ *** | $3.74\times {10}^{-5}$ *** | $4.10\times {10}^{-5}$ *** | $3.95\times {10}^{-5}$ *** | $3.81\times {10}^{-5}$ *** |

(0.1, 0.5) | $1.98\times {10}^{-4}$ *** | $1.44\times {10}^{-4}$ *** | $6.00\times {10}^{-5}$ * | |||

(0.5, 1) | $3.70\times {10}^{-4}$ *** | $3.03\times {10}^{-4}$ *** | $3.18\times {10}^{-4}$ *** | $2.04\times {10}^{-4}$ ** | ||

(1, 5) | $2.67\times {10}^{-4}$ *** | $3.15\times {10}^{-4}$ *** | $2.64\times {10}^{-4}$ *** | $2.27\times {10}^{-4}$ *** | $2.66\times {10}^{-4}$ *** | $2.07\times {10}^{-4}$ *** |

(5, 10) | $8.91\times {10}^{-4}$ *** | $6.82\times {10}^{-4}$ *** | $8.26\times {10}^{-4}$ *** | $8.18\times {10}^{-4}$ *** | $6.77\times {10}^{-4}$ *** | $5.85\times {10}^{-4}$ *** |

(10, 25) | $3.63\times {10}^{-4}$ *** | $2.32\times {10}^{-4}$ * | ||||

(500, 1000) | $-3.22\times {10}^{-4}$ * | $-3.85\times {10}^{-4}$ * | ||||

(1000, 5000) | $-4.67\times {10}^{-4}$ * | |||||

(5000, 10,000) | $-8.39\times {10}^{-4}$ * | $-7.49\times {10}^{-4}$ * |

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |

© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Zhang, Y.; Garg, R.; Golden, L.L.; Brockett, P.L.; Sharma, A.
Segmenting Bitcoin Transactions for Price Movement Prediction. *J. Risk Financial Manag.* **2024**, *17*, 128.
https://doi.org/10.3390/jrfm17030128

**AMA Style**

Zhang Y, Garg R, Golden LL, Brockett PL, Sharma A.
Segmenting Bitcoin Transactions for Price Movement Prediction. *Journal of Risk and Financial Management*. 2024; 17(3):128.
https://doi.org/10.3390/jrfm17030128

**Chicago/Turabian Style**

Zhang, Yuxin, Rajiv Garg, Linda L. Golden, Patrick L. Brockett, and Ajit Sharma.
2024. "Segmenting Bitcoin Transactions for Price Movement Prediction" *Journal of Risk and Financial Management* 17, no. 3: 128.
https://doi.org/10.3390/jrfm17030128