Complete Settings List of BelkaMiner EA (v1.15
February 2020)


Modes


  • SystemMode – operating modes: ‘Machine Learning is used for data
    mining and machine learning, ‘BackTesting & Trading is used for backtesting, researching and
    live trading.


Money Management

  • LotSize – fixed lot size for trading if UseMargin% and Balance are
    set to 0.
  • Balance – proportional lot size calculated by the formula: LotSize * AccountBalance/
    Balance. It works together with LotSize. This MM proportionally increases (decreases) lot sizes
    when the account balance increases (decreases), respectively. The smaller the value, the higher the risk. In case of non-zero value,
    the EA calculates lot sizes based on the 
    account balance, e.g., 0.05 lots for every $1000. If ‘Balance‘ set to a negative value, then the EA
    calculates lot sizes based on the account equity.
  • MaxRiskPerTrade% – maximum risk (as a percentage of the account balance) per trade based on
    the initial StopLoss value. If MaxRiskPerTrade%=0, the EA will use LotSize or UseMargin%.
  • MaxLot_Overall – maximum lot size that the EA can use for trading.



GMT


  • AutoGMT_Detection – if true, the EA will automatically detect your broker’s time zone in
    live trading
    . To use this feature, you must allow requests to:
    https://www.worldtimeserver.com
    https://belkaglazer.com (backup URL to determine GMT offset)


GMT[m] – GMT has been calculated manually, 
GMT[a] – GMT has been detected automatically.
+3 – current GMT offset (DST is taken into account), 
[NY] – NY Close Broker.

  • NYCloseBroker – set this parameter to false if your broker’s server time is
    not set to New York Сlose. If you use a NY close broker (who closes daily candles at 5 pm New York time, that corresponds
    to GMT+2 in winter and GMT+3 in summer), then no need to change the default settings.
  • GMT_offset – GMT offset of your broker in winter. This parameter should be configured only if NYCloseBroker=false.
  • Daylight_Saving_Time – set this parameter to US/Europe if
    your broker changes the GMT offset in the summer period. This parameter should be configured only if NYCloseBroker=false.


Execution

  • MaxSpreadPip – if the spread exceeds this value, then the EA will not allow opening a new trade. It
    works only with a market order, pending orders will not be closed.

  • MaxSpreadToExitPip – if the spread is above this value, then the EA will not allow
    closing an open trade.
  • MaxSlippagePip – parameter allows you to specify the maximum slippage (as a max. deviation from the
    quoted price) for a market order. It works only with Instant Execution (for example, standard,
    classic, micro, and nano accounts). This parameter has no effect on an ECN account.
    If MaxSlippagePip > 0, the EA will control slippage using the maximum deviation from the ask/bid price. If a broker
    is unable to fill an order within a specified deviation, then the order will be rejected.



Time

  • OrderHourStart/Stop – hour when the EA should start/stop looking
    for trading signals.
  • OrderDayOfWeek – day(s) of the week when the EA is allowed to open a
    new trade.

  • DaysOffBeforeAfterNY – a trading pause which is activated before/after the New Year for
    a specified number of days. For example, if DaysOffBeforeAfterNY=3, then the EA will not open new trades from
    December 29 to January 3.



Basic Entry Rules

  • StrategyType – a type of ‘White-box’ algorithm. The
    algorithm is based on boolean logic, open and clear rules. It is used to generate entry points and gives many trades with a low average
    trade. It must have a ‘trading edge’ that may be too small to cover trading costs. The edge can only be seen after the spread/commission is
    turned off or set to a minimum value.
     
    Reversal (PCh) is a mean reversion strategy that uses reversals
    between support and resistance levels.
     
    Breakout (PCh) is a breakout strategy that enters the market when the price moves outside a
    defined price range (support and resistance levels).
     
    Momentum (PA) is a momentum strategy that uses significant price movements in one direction.
  • DailyATR_Period – period of DailyATR. It is used to
    normalize the volatility.

  • PCh_Type –  methods used for calculating the Price Channel:
     Donchian
    Channel
     is formed by taking the highest high and the lowest low of the last n periods. The area
    between the high and the low is the channel for the period chosen.

     Bollinger
    Bands
     – is based on two standard deviations away from a simple moving average.

  • PCh_Period – a period of the Price Channel.
  • PCh_OffsetPip – offset of the Price Channel levels.
  • PA_Bars – number of bars to analyze price movement (for Momentum strategy).
  • PA_Size%ATR – minimal size (in %ATR) of price movement (for Momentum strategy).



Basic Exit Rules

  • StopLoss, TakeProfit – stop loss and take profit. They can be
    expressed in pips (in case of value > 3) or %ATR (in case of 0
    < value <3
    ). If value = 0, the EA places SL(TP) in the middle of the price range (HL/2 level).
  • StopBar – time-based stop loss that automatically closes trades after a certain number of
    bars, regardless of other conditions (0 = off).
  • FridayStopHour – hour to close trades on Friday (-1 = off).
  • TrailingStopPip – in case of value > 0: trailing stop in
    pips
     when a position is in profit, in case of value < 0: minimum profit level in pips when
    SL should be moved to breakeven (0 = off).

  • TrailingActivationPip – determines the number of pips to activate ‘TrailingStopPip‘.



Machine Learning


  • Use_Machine_Learning – enable/disable ML in backtesting and live trading. If disabled, the
    EA will only use rules of the basic strategy.

  • Machine_Learning_Technique – ML techniques:
     Clustering
    (Unsupervised)
     – the EA groups a set of entry points (dataset) in such a way that points in the same group
    (called a cluster) are more similar to each other than to those in other groups (clusters). Then we test each cluster and try to find
    profitable ones with good performance. We assume the points in the profitable clusters have certain properties that make them
    profitable. Clustering is a main task of exploratory data mining, and a common technique for statistical data analysis, used in many
    fields.






     
    Neural Network (Supervised) – the EA trains 
    perceptron using
    raw data without clustering.




  • ConfigFile – configuration file containing the results of clustering and machine learning. It is located in
    the ‘CommonFiles‘ directory. The default config file (for the default strategy) is created after the
    first launch of the EA. Read
    more ->



Input Data / Metrics


  • UseTradingHours_Measure – use the trading hour as a measure. It can be used for Intraday
    seasonality
     detection.

  • UseEMA_Measure – daily EMAbased metrics normalized by ATR and
    calculated by the formula: (Close-EMA(n)) / ATR(m). It can be used for trend detection.

  • EMA_Period – perion of the daily EMA.

  • UseMomentum_Measure – price movement before the entry point
    normalized by ATR. It may help figure out how the market reacts to certain price movements.

  • Momentum_Bars – number of bars to calculate the price
    movement.

  • UseDailyVolatility_Measure – volatility-based metrics calculated by the formula:
    ATR / ATR(30). It may help in Daily volatility filtering.

  • UseIntradayVolatility_Measure  volatility-based metrics normalized by ATR and
    calculated by the formula: (Upper PCh level – Lower PCh level ) / ATR. It
    may help in Intraday volatility filtering.

  • UseHurst_Measure  hurst-based
    metrics. It is used as a measure of memory of time series.

  • Hurst_Bars – number of bars to calculate Hurst exponent.



Clustering settings


  • ClusteringMethod – method used for clustering:
     K-means++
    is one of the simplest and popular unsupervised machine learning algorithms with choosing
    the initial values
    .

  • NumberOfClusters – desired number of clusters for k-means clustering. 30….50 is
    recommended. The optimal number depends on the size of the dataset (number of points/vectors).

  • NumberOfRestarts – max. number of restarts to find a good / the global optimum. Try to
    reduce it if clustering takes too long.

  • OutlierDataPercentile – parameter for calculating 2 percentiles (min
    & max) to filter
    outliers
     before normalizationMin percentile = OutlierDataPercentile, Max percentile
    = 100-OutlierDataPercentile.


  • NormalizationMethod
     – adjusting values measured on different scales to a notionally
    common scale:
     Z-Score uses mean and standard
    deviation.
     It is calculated as: z = (x-mean) / stdev.

  • WhatToSave – data to be saved in the Config
    file
    :
    – 
    DoNotSavePoints – the EA will only save mean, stdev and coordinates of centroids. Fast, but NN (Neural
    Network
    ) training (to classify the cluster number) is not possible in the
    future.
    – 
    SavePointsOnly – the EA will save mean, stdev, coordinates of centroids and points with cluster numbers.
    Fast, NN can be trained later.
    – 
    SavePoints & TrainPerceptron – the EA will save everything and train NN. This may take a
    long time
     (hours or even days, it depends on the settings).



Perceptron Settings

Multilayer perceptron



  • NeuronInHiddenLayer 
    – number of neurons in the hidden
    layer
    . If zero, then no hidden layer will be used.

  • TrainingAlgorithm – training algorithm:
     Levenberg–Marquardt algorithm
    (LMA or just LM), also known as the Damped least-squares (DLS) method, is used to solve non-linear least squares problems.
     L-BFGS is
    an optimization algorithm in the family of quasi-Newton methods that approximates the Broyden–Fletcher–Goldfarb–Shanno
    algorithm (BFGS) using a limited amount of computer memory. It is a popular algorithm for parameter estimation in machine learning.

  • NumberOfRestarts – max. number of restarts to find a good optimum. Try
    to reduce it if learning takes too long.



Clustering Trading Setup


  • Classificator – method used to classify the cluster number in backtesting and live
    trading
    :
     Euclidean
    Distance
     uses the shortest distance to the centroids calculated by the Pythagorean
    formula
    . Fast but less accurate method.

    – 
    Neural Network – train NN to classify the cluster number. This
    may take a long time
     (hours or even days, it depends on the settings).
    – 
    Saved Neural Network uses NN saved in the Config file. Fast if NN is saved, if not – the EA will train NN and save results to
    the Config file.

  • ClusterNumberForOptToBuy(Sell) – parameters to find stable and profitable clusters.

  • ClusterNumber(s)ToBuy(Sell) – cluster numbers to trade, separated by a comma.



Order setting

  • UserComment – user comment used to label orders opened by the EA.
  • MagicNumber – a unique basic identifier of EA’s orders. This
    number should be less than 99999.

  • Graphics – this option allows you to enable/disable graphic objects
    of the EA on a chart.



Source link

Register at Binance