Amrullah Deep Liquidity
- This article was considered for deletion at Wikipedia on July 9 2019. This is a backup of Wikipedia:Amrullah_Deep_Liquidity. All of its AfDs can be found at Wikipedia:Special:PrefixIndex/Wikipedia:Articles_for_deletion/Amrullah_Deep_Liquidity, the first at Wikipedia:Wikipedia:Articles_for_deletion/Amrullah_Deep_Liquidity.
Amrullah Deep Liquidity (ADL) is a high performance technical indicator that is built using deep learning artificial intelligence (AI). ADL is a type of statistical method that finds the ideal trades when trading a cryptocurrency. ADL, designed by Muhd Amrullah, tells traders when to take a trade and how much to put in a trade. ADL can do this because it can learn how any market maker move in any chosen market.
Arrows and Indicators
On a Tradingview chart, ADL will display an arrow that tells you when to enter a trade. Traders can also see the amount to trade beside the arrow.
Backtesting is a robust way to measure performance. The average 12-month percentage net profit of ADL backtested across BTCUSD is 1621% and the average 1-month percentage net profit of ADL backtested across BTCUSD is 135% during the periods between 2017 to 2019.
Across a 5-year period for BTCUSD, the Profit Factor (PF), a standard used commonly by traders to measure effectiveness of a strategy, measures to 8.414 . Across a similar period for ETHUSD, the PF measures to 28.89. A PF that is above 3.5 is considered world-class by traders. Traders can see choose a different pair such as ETHBTC, BCHBTC or BNBETH or any other trading pairs that fit their appetite or have the Profit Factor they are optimizing for. Traders can also select a different timeframe such as 1H. The green equity curve increases as the trade takes profits.
ADL can manage the risks traders take in every trade but does not guarantee profits. The use of Amrullah Deep Liquidity may vary widely among traders. The factors largely depend on the trader's discipline, the type of cryptocurrency the trader has chosen and the preferred time frame the trader has chosen.
Various analysis of the effectiveness of ADL with other indicators have been performed. The building of ADL took two years to complete from 2017 to 2019 and involves a process of analyzing terabytes of data, pre-processing of data, designing of deep learning models, using methods such as ensemble learning. For processing time-series data such as cryptocurrency price feeds, Tensorflow and Keras are used. For analysis of the data, Seaborn and the Scipy family of tools are most effective. Scikit and Xgboost are handy whenever required.