• Visão global
  • Comentários (2)
  • Discussão (51)

HFT Deep Learning

Artificial Neural Networks – High-Frequency Trading

Modern techniques like artificial neural networks (ANN) are best used for high-frequency trading for several reasons. First, they mimic human intelligence but they mostly don’t reach a human’s level of intelligence, therefore, there is no point in using those techniques on a time scale at which a human could easily be working. Their advantage comes from the speed of operation and constant activity. Second, we need a lot of data to train neural networks efficiently and this amount of data will only be found in high-frequency trading. Forex has all in all quite few instruments with limited relevant past data on the daily or weekly time-scale.

Furthermore, High-frequency trading is a type of scalping strategy where we identify noise around the true value of the instrument. This is different from long-term trading that attempts to follow meaningful movements of the instrument according to fundamental analysis. A good time-scale to work on is the minute time-scale. This time-scale is full of noise which will be captured by the algorithm in order to sell and buy during the strong movements of the market. That is why it is irrelevant to take old data to predict these movements over a period of one minute.


To improve the performance and robustness of HFTs, the following issues need to be addressed:

  1. Message latency:
    To achieve message latency, the underlying network should be made of high-performance computing elements. Low latency networks are critical for timely message passing. A lot of brokers are providing co-location facilities to minimize network latency. Improving the message latency will also reduce the time taken between the decision making and execution.
  2. High-performance computing:
    A high-performance computing setup is vital for HFTs. This will enable the ability to price and calculate risks and positions at portfolio scale in near real time. It will also support the ability to analyze data and performance after the execution to identify and evaluate market strategies thus creating a platform for informed decision making. This is a primary requirement of HFT.


Deposit: $100

Inputs Description

  • Take Profit
  • Stop Loss
  • Fixed Lot Size
  • Use Money Management
  • Risk Percent - Risk per trade
  • Candle Size - Minimum candle size for entry in points
  • Magic Number
  • Expert Comment
  • Max Spread In Points
  • Trailing Stop
  • MA Period - Moving average period
  • MA Method - Moving average Method
  • RSI Period
  • Limit Trade Per Candle - Use max number of trades per candle
  • Max Trades Per Candle - Max number of trades per candle
  • Close Trade At End Of Candle
  • Send Push Notification on Entry
  • Hide SL/TP From Your Broker
  • Weight 0... Weight 9 - The weights in an artificial neural network are an approximation of multiple processes combined that take place in biological neurons
  • Trade on Monday
  • Trade on Tuesday
  • Trade on Wednesday
  • Trade on Thursday
  • Trade on Friday
  • Using Trade Hour
  • Start Hour - Start hour for trading
  • End Hour - End hour for trading

The default Time Frame is M1 - one minute

Anthony Stone
Anthony Stone 2018.08.31 18:48 

We had a slow start but this is now becoming another good addition to my portfolio. A solid 4 stars from me at the moment at this early stage. If performance continues to impress then i'll upgrade to 5 starts.

Update 04/01/19 - I switched this EA off a good few months ago as there were consistent losses taking place. Untill an update comes along i can't recommend trading with this EA at present. Hopefully Anddy will get round to this update soon and i can updated my findings

Alex_B 2018.07.26 10:05 

Running HFT since 26.07.2018 on my VPS with authors set files, each with its own unique magic number.

Total profit so far until 7th of August 2018: -35,71% !!!

Please see also comment #32 and my live signal on HFT (see link in comment #33). My original review was deleted by MQL5 as it included the myfxbook Link...

If it turns out that this EA will get profitable I will of course update my rating.

Update 21.08.2018: Currently it looks as HFT is recovering from it' starting losses. Rating therefore raised to 3 stars

Update 22.10.2018: HFT is unfortunatelly not performing well. Rating lowered to 2 stars