In this research we make the first attempt to construct automated Bitcoin trading systems that are based on classical candlesticks patterns. We than tried to alter the classical formations for better trading results. Our data consists Bitcoins prices from the beginning of 2012 till the end of July 2020. WE found that that out of the tree classical candlesticks pattern only Engulfing has been fertile in predicting Bitcoin's price trends shifts. The classical Engulfing pattern generated Profit Factor (PF) of 3.54 and $38,349 Net Profit (NP). We also found that using a strength proxy of 0.9% may improve the classical pattern results. The classical Harami formation failed to produce positive trading results and therefore does not fit Bitcoin trading platforms. On the other hand, a reversed Harami was proven to be a fertile Bitcoin trading strategy. Our research also finds that classical four bars Kicker signals seldom appears and therefore cannot help Bitcoin traders. On the other hand, a reversed Kicker pattern has found to be a winning strategy with relatively low risk. This strategy fits particularly long positions with 74.36% Percent of Profitable (PP) trades and 6.92 PF. We also find that all the examined candlesticks patterns better predict long trends than short trends.

在这项研究中，我们首次尝试构建基于经典烛台模式的自动比特币交易系统。然后，我们尝试更改经典形式，以获得更好的交易结果。我们的数据包括从2012年初到2020年7月底的比特币价格。我们发现，从树上的经典烛台模式来看，只有Engulfing在预测比特币价格趋势变化方面具有丰富的经验。经典的吞没模式产生的利润因子（PF）为3.54，净利润（NP）为$ 38,349。我们还发现使用0.9％的强度替代可以改善经典图案结果。经典的Harami格式未能产生积极的交易结果，因此不适合比特币交易平台。另一方面，反向Harami被证明是一种肥沃的比特币交易策略。我们的研究还发现，很少出现经典的四格跳动信号，因此无法帮助比特币交易者。另一方面，反向的踢球者模式被认为是风险相对较低的制胜法宝。该策略特别适合做多头寸，获利（PP）交易的百分比为74.36％，PF为6.92。我们还发现，所有检查的烛台形态比短期趋势更好地预测了长期趋势。