• Quick Search:
Unlike other financial networks, such as stock and currency trading, blockchain-based cryptocurrencies have the entire transaction graph accessible to the public (i.e., all transactions can be downloaded and analyzed). A natural question is then to ask whether the dynamics of the transaction graph impact the price of the underlying cryptocurrency. Analyzing transactions and addresses to track the Bitcoin economy has become an important research direction. Blockchain features, such as average transaction amount, are shown to exhibit mixed performance for cryptocurrency price forecasting.

Bitcoin Price Prediction

We provide Bitcoin price data downloaded from coinmarketcap.com since 2011 (when BTC was first traded on online exchanges). Data consists of five columns

date,price,year,day,totaltx
6/1/2011 0:00,9.57,2011,152,6464
6/2/2011 0:00,10.6,2011,153,7632
6/3/2011 0:00,14.29,2011,154,8502
6/4/2011 0:00,18.89,2011,155,8766
6/5/2011 0:00,16.7,2011,156,8675

Task: Temporal prediction - Bitcoin price prediction by using signals (e.g., chainlets) in the Bitcoin transaction network.

Challenge: Bitcoin transaction network activity may have a delayed impact on the Bitcoion price. For example, increased transaction counts may draw the attention of users and cause price changes in a later time. In our research works (e.g., Chainnet), we found that the impact of increased activity manifests itself with a delay of 3-7 days. Furthermore, traditional network metrics are mostly powerless in this prediction task. For example, mean degree of addresses, number of new addresses, mean coin amount transferred in transactions and address network average clustering coefficient had no predictive power. We found that price and total number of transactions were useful predictors in Chainnet models.

Prediction target attribute: price

Full Dataset

See the price data in a csv file.

Cite Our Dataset:

	@inproceedings{chartalistNeurips2022,
  author    = {Kiarash Shamsi and Yulia R. Gel and  Murat Kantarcioglu and Cuneyt G. Akcora},
  title     = {Chartalist: Labeled Graph Datasets for UTXO and Account-based Blockchains},
  booktitle = {Advances in Neural Information Processing Systems 36: Annual Conference
               on Neural Information Processing Systems 2022, NeurIPS 2022, November 29-December
               1, 2022, New Orleans, LA, USA},
  pages     = {1--14},
  year      = {2022},
  url       = {https://openreview.net/pdf?id=10iA3OowAV3}
  }

Baseline

Our novel concept of k-chainlets on Bitcoin that expands the ideas of motifs and graphlets to Blockchain graphs. Chainlet analysis provides a deeper insight into local topological properties of the Blockchain and the role of those local higher-order topologies in the Bitcoin price formation. We found that certain types of chainlets have a high predictive utility for Bitcoin prices. Furthermore, extreme chainlets exhibit an important role in the Bitcoin price prediction.

Baseline Reference Paper