This research focuses on analyzing the relationship between Bitcoin prices, trading rates and volume to predict prices in 15-minute candle intervals. The study proposes the use of Long Short-Term Memory (LSTM) neural network for developing a reliable prediction model. Historical Bitcoin data was obtained from the Binance API and underwent collection and preprocessing. The performance of the prediction model is evaluated against the real one. The findings contribute to the analysis of cryptocurrency markets, enabling investors to make informed real-time decisions, facilitating the development of advanced trading algorithms and risk management strategies in the cryptocurrency space.
An implementation of a LSTM on bitcoin chart.
Price, trading rates, and volume of Binance transactions for Bitcoin pairs in 2020, with a 15-minute interval.
Milliseconds time from 2017-31-12 to 2023-01-06.
Volume from Bitcoin to altcoins, memecoins or NFTs.
Volume from Bitcoin to FIAT or Stable coins.
Volume from Bitcoin to altcoins, memecoins or NFTs.
Trades from Bitcoin to FIAT or Stable coins.
Price in USDT
The crypto market operates on cycles driven by FOMO (Fear of Missing Out) and FUD (Fear, Uncertainty, and Doubt). Every four years, Bitcoin undergoes a halving, reducing the mined supply by half. This dataset focuses on the 2020 halving, providing insights into the flow of money during this period.
- Integrated binance API module
- Native Threads support on request
- Easy to use and large documentation
- Http remaker (personalized HTTP response)
This code is distributed under the GPL-3.0 license. See the LICENSE file for more information.