This research is the first attempt to create Machine Learning (ML) algorithmic systems that would be able to intraday trade automatically popular cryptocurrencies using oscillators that are commonly used to trade other financial assets. It uses intraday price data of Bitcoin, Ethereum, Binance Coin, Cardano, and XRP with different trading time frames that vary from 5 to 180 min. Our results show that the RSI system is the best algorithmic trading system for cryptocurrency intraday trading. The RSI-based system has out beat the B&H strategy for all five cryptocurrencies. Moreover, it has proven its ability to improve trading performances under any market conditions up or down trends. The other two trading systems that were based on the MACD and Keltner Channels oscillators were found to outperform the B&H strategy for Bitcoin, Binance Coin, and XRP while it was beaten by the B&H strategy for Ethereum and Cardano. Although cryptocurrencies are known for their high volatility, this research has proven that longer time frames such as 60 and 120 min have produced better trading results than shorter time frames such as 5 and 15 min.

这项研究是首次尝试创建机器学习 (ML) 算法系统，该系统能够使用通常用于交易其他金融资产的振荡器自动在日内交易流行的加密货币。它使用比特币、以太坊、币安币、Cardano 和 XRP 的日内价格数据，交易时间范围从 5 分钟到 180 分钟不等。我们的结果表明，RSI 系统是加密货币日内交易的最佳算法交易系统。基于 RSI 的系统已经击败了所有五种加密货币的 B&H 策略。此外，它已证明其能够在任何市场条件上升或下降趋势下改善交易表现。其他两个基于 MACD 和 Keltner Channels 振荡器的交易系统被发现优于比特币 B&H 策略，Binance Coin，和 XRP，而它被以太坊和卡尔达诺的 B&H 策略击败。尽管加密货币以其高波动性着称，但这项研究证明，较长的时间范围（例如 60 分钟和 120 分钟）比较短的时间范围（例如 5 分钟和 15 分钟）产生更好的交易结果。