Paper: arXiv 2403.00770

Authors: Juan C. King, Roberto Dale, José M. Amigó

Abstract

The objective of this paper is the construction of new indicators that can be useful to operate in the cryptocurrency market. These indicators are based on public data obtained from the blockchain network, specifically from the nodes that make up Bitcoin mining. Therefore, our analysis is unique to that network. The results obtained with numerical simulations of algorithmic trading and prediction via statistical models and Machine Learning demonstrate the importance of variables such as the hash rate, the difficulty of mining or the cost per transaction when it comes to trade Bitcoin assets or predict the direction of price. Variables obtained from the blockchain network will be called here blockchain metrics. The corresponding indicators (inspired by the “Hash Ribbon”) perform well in locating buy signals. From our results, we conclude that such blockchain indicators allow obtaining information with a statistical advantage in the highly volatile cryptocurrency market.

Complexity vs Empirical Score

  • Math Complexity: 4.0/10
  • Empirical Rigor: 6.5/10
  • Quadrant: Street Traders — empirical and applied, lower math complexity

Why this score: The paper employs relatively standard statistical and machine learning models (Random Forest, LSTM) without deep mathematical derivations, but heavily emphasizes empirical validation through numerical simulations, algorithmic trading backtests, and specific data handling (Quandl platform) with available Python code.

Research Flowchart

  flowchart TD
  A["Research Goal:<br>Construct new crypto trading indicators<br>using Blockchain Metrics"] --> B["Data Collection"]
  
  B --> C{"Bitcoin Blockchain Data"}
  C --> C1["Hash Rate"]
  C --> C2["Mining Difficulty"]
  C --> C3["Transaction Cost"]
  
  B --> D["Methodology"]
  D --> D1["Algorithmic Trading Simulations"]
  D --> D2["Statistical Models"]
  D --> D3["Machine Learning"]
  
  D1 & D2 & D3 --> E["Computational Process:<br>Develop 'Hash Ribbon' inspired<br>indicators & test efficacy"]
  
  E --> F["Key Findings/Outcomes"]
  F --> F1["Indicators provide<br>statistical advantage"]
  F --> F2["Identify buy signals<br>effectively"]
  F --> F3["Hash Rate, Difficulty, &<br>Cost are significant variables"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes new blockchain‑based technical indicators called blockchain ribbons, extending the Hash Ribbon methodology with additional on‑chain metrics.

Source sentence
The result are new indicators called blockchain ribbons.

Assumes 4

  • The Quandl public REST API provides accurate daily on‑chain metric data for Bitcoin. data
    Source sentence
    After surveying data from the Bitcoin blockchain network, we selected datasets obtained through Quandl; we chose all the datasets available on that platform belonging to the mining of the Bitcoin network.
  • Trading simulations use a flat transaction fee of about 1% per trade. execution
    Source sentence
    Trading fees are around 1% in the cryptocurrency market.
  • Long and short signals are generated by crossings of 30‑day and 60‑day moving averages. model
    Source sentence
    moving averages of 30 and 60 days were applied to obtain long and short signals based on the crossings of the moving averages.
  • Historical price and metric series start on August 16th, 2010 and are representative of Bitcoin market behavior. data
    Source sentence
    the time series that we analyzed start on August 16th, 2010 (t = 1) because the Bitcoin price was ...

Data

  • Median Transaction Confirmation Time (ATRCT) · Quandl crypto-onchain public
    2009-10-03 · Bitcoin blockchain
    Source sentence
    (1) Median Transaction Confirmation Time (ATRCT); (2) Average Block Size (AVBLS); (3) Api Block-Chain Size (BLCHS)

Recognised sources: Nasdaq Data Link (Quandl)

Costs

Modeled Costs are part of the reported headline results.

Trading fees are around 1% per trade and are incorporated in the Total Profit calculation.

Source sentence
Trading fees are around 1% in the cryptocurrency market.

Finds 5

  • The CPTRA metric records the best performance in both long and short operations during the December 2017‑July 2022 period. performance
    Source sentence
    It is also worth highlighting that the CPTRA metric records the best performance in both long and short operations in the period December 2017 - July 2022.
  • Blockchain metrics increase profitability in long operations, with four metrics (NADDU, CPTRA, TRFUS, TRFEE) outperforming the benchmark Bitcoin price. performance
    Source sentence
    The four metrics NADDU, CPTRA, TRFUS and TRFEE outperform the benchmark MKPRU (Bitcoin price).
  • LSTM with percentages achieves low MASE values (0.74 test, 0.63 train), indicating good predictive learning. predictive-accuracy
    Source sentence
    the lowest values of the MASE metric are obtained with the algorithm “LSTM with percentages”, namely: 0.74 using testing data and 0.63 using training data.
  • The Adjusted CPTRA ribbon (AdCPTRA) obtains the best performance in short‑signal algorithmic trading and is the most important variable in the Random Forest model. performance
    Source sentence
    it is worth highlighting the AdCPTRA indicator, which is the one that obtained the best performance in short signals in algorithmic trading systems and also the most important variable in the Random Forest algorithm, exceeding the value of the second most important variable by a factor of 3.
  • Some blockchain metrics (e.g., ETRAV) have low correlation with price yet still perform well in trading and predictive models, showing exceptions to the general rule. other
    Source sentence
    For example, ETRAV has the lowest coefficient of correlation with the Bitcoin price and yet performs well in algorithmic trading and is also an important variable in predictive models.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 20 related papers). Coverage is partial, so this is neither replication nor refutation.

Cited by 1 paper in the archive

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.

Cited References (35)

  1. Cashing in on the Dow: using Dow theory to trade and determine trends in today’s markets (2020)
  2. Dow theory for the 21st century: Technical indicators for improving your investment results (2008)
  3. The wave principle (2019)
  4. Optimal negative weight moving average for stock price series smoothing (2014)
  5. Bitcoin whitepaper (2008)
  6. Trading with Oscillators: Pinpointing Market Extremes–theory and Practice (1997)
  7. The Richard D. Wyckoff Method Of Trading And Investing In Stocks (1931)
  8. El m 'e todo Wyckoff: Claves para entender los fundamentos del trading (2016)
and 27 more
  1. Using bollinger bands (1992)
  2. Market timing with moving averages: The anatomy and performance of trading rules (2017)
  3. Exponential Movil Average Ribbon (2019)
  4. Trading with Ichimoku clouds: the essential guide to Ichimoku Kinko Hyo technical analysis (2010)
  5. A brief survey of cryptocurrency systems (2016)
  6. Analysis of a SHA-256 variant (2005)
  7. Hash Ribbon (2020)
  8. Bitcoin mining acceleration and performance quantification (2014)
  9. Technical analysis of stock trends (2018)
  10. Handbook of Applied Cryptography (1997)
  11. A new coefficient of correlation (2021)
  12. Deep learning for stock market prediction (2020)
  13. An on-chain analysis-based approach to predict ethereum prices (2021)
  14. Learning and predictability via technical analysis: evidence from Bitcoin and stocks with hard-to-value fundamentals (2021)
  15. Does the hashrate affect the bitcoin price? (2020)
  16. Pick a nonce and try a hash (2019)
  17. Nadsaq Official Web (2009)
  18. Personal Github account of Juan Carlos King Perez (2023)
  19. Impact of Currency Pairs, Time Frames and Technical Indicators On Trading Profit in Forex Spot Market. (2009)
  20. A study on technical indicators in stock price movement prediction using decision tree algorithms (2016)
  21. Understanding blockchain technology (2018)
  22. Analysis of hashrate-based double spending (2014) — link
  23. Efficient market hypothesis: A review of theory and empirical work (1970)
  24. The adaptive markets hypothesis: Market efficiency from an evolutionary perspective (2004)
  25. Alpha trading: profitable strategies that remove directional risk (2011)
  26. Deep learning for stock market trading: a superior trading strategy? (2019)
  27. A multi-layer and multi-ensemble stock trader using deep learning and deep reinforcement learning (2021)

0 of 35 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.