SOMFX1Predictor

SOMFX1Predictor

If you are trading using candlestick shapes and want to improve their methods of modern technology, this indicator is for you. In fact, it is part of a set of tools, which is based on neural network engine SOM (Self-Organizing Map, SOM) for the detection and prediction of candle shapes, as well as for the study of input data and results of operation of the network. The kit contains:

  • SOMFX1Builder – script to train the neural networks; it creates a file with the generalized data of the most characteristic price figures, which can be used to predict the bars in the figures or in a separate window (via SOMFX1 indicator) or directly on the main graph (SOMFX1Predictor);
  • SOMFX1 – an indicator for the prediction and visual analysis of price figures, the input and output of the trained neural network (in a separate window);
  • SOMFX1Predictor – the indicator to predict price patterns directly in the main window;

Tools implemented separately from each other due to the fact that MetaTrader 4 has some limitations, e.g., at the moment it is impossible to run lengthy calculations in the indicator, since LEDs are executed in the main stream.

In short, the whole process of price analysis, network training, recognition of figures and their predictions are lies in the following steps:

  1. Creating a neural network using SOMFX1Builder;
  2. Analysis of the quality of the received network via SOMFX1; If unsatisfactory, return to step 1 to the new settings; you can skip step 2 if desired;
  3. Using the final version of the network to predict the shapes using SOMFX1Predictor.

Details of the steps 1 and 2 – neural network learning and visual data analysis – examines the pages of the relevant instruments – SOMFX1Builder and SOMFX1. This document describes how to use SOMFX1Predictor.

Warning: requires light for a file, create a script SOMFX1Builder. Thus, you must either download SOMFX1Builder and generate a file according to your requirements or ask a friend (having a script) to create a file for you.

Work principles

This indicator – the simplest part of the neural network tool. He takes the file name of the trained neural network and shows the forecast on the main chart. Prediction depicted red line, which may be partly thick and partly thin. Thick pieces means the place where the network is more confident in the forecast. Predicting the start of the vertical dotted line yellow color called “FromPastToFuture”. 
You can move the line to an arbitrary position, and the indicator will read the sample rates near this place, to submit it to the SOM input and display the prediction of the next price movement. Current sample price – a fragment of the price increments taken from PatternSize bars to the left of the line “FromPastToFuture”. When you open the display for the first time, the line is placed on the bar LearnStop. If you move the line on the 0th bar, the indicator will automatically shift it to every new 0th bar as they become available, then there is a line “glued” to the last bar.

Note that the prediction starts with the current bar, that is, the current bar is also predicted. This is done because the current bar is usually not complete and requires the prediction itself. Moreover, if we take another nazakonchenny bar into account, the prediction would be contradictory (unstable at the time of the formation of the bar). Due to the fact that the current bar is also predicted by the red line starts at 1 bar to the left of the vertical line “FromPastToFuture” – it shows how the price should be changed on the current bar.

The indicator may optionally output the neural network itself (map) if ShowMap to true. In this case, the chart there are two square card.

Left quadrant shows the current activity of neurons neurons with higher excitation displayed in red, and with a smaller – blue. In other words, the red box on the map, the better the neuron corresponds to the current price figure.

Second – right – square displays a map of “population density” cells, that is, the color indicates the number of samples in the input data, which fell into the appropriate cell and formed her price pattern. Red indicates a high density, and the blue – low. Gray cells do not have the corresponding samples in the input data.

Options

  • LearnStart – bar number in the history of where to start training data, or the exact date and time of the bar (in the format “YYYY.MM.DD HH: MM”); This parameter – a string that allows you to type and number, and the date; This parameter is used here not to teach, but to re-create the training set of data (corresponding to the neural network), which is important if the parameter UseAverage is true (see below.), and also for visualization “densities”; ; 5001 – default if you enter the auto-generated name in the parameter NetFileName (See. Below), shares the name of the indicator into components and use them instead of the other parameters, including LearnStart; In other words, this option does not affect the work, if the parameter  NetFileName entered automatically generated name of the neural network file;
  • LearnStop – bar number in history, where the training data ends, or the exact date and time of the bar (in the format “YYYY.MM.DD HH: MM”); This parameter – a string; ; 1 – Default This parameter is used here not to teach, but to re-create the training set of data (corresponding to the neural network), which is important if the parameter UseAverage is true (see below.), and also for visualization “densities”; This option does not affect the operation, if the parameter  NetFileName entered automatically generated name of the neural network file;
  • PatternSize – the number of bars in the same figure; default – 5; This option does not affect the operation, if the parameter  NetFileName entered automatically generated name of the neural network file;
  • GridSize –  card size; is the number of cells / neurons vertically and horizontally; allowable value: 3 – 50; default – 7; This option does not affect the operation, if the parameter  NetFileName entered automatically generated name of the neural network file;
  • PredictionBars – the number of prediction bars; ; 10 – by default Please note that the next bar is predicted with less precision than the previous one, since the prediction errors accumulate;
  • UseAverage – This switch is a special regime; when it is disabled (false, the default) predictions are made on the basis of the weights of the winning neuron, that is, they are determined solely by the card; when the mode is enabled (true), predictions are made on the basis of the average prices of all samples are displayed on the winning neuron; this means that the prediction is involved not only a map, but also the training sample data, which is why it is important to specify the exact date and time for training segment; with this option you can “play” to get the best results: the use of weights of neurons – is a classic approach, but the use of average prices brings an additional binding to the raw data – in particular, in this case it is possible to take into account the dispersion and thereby evaluate the prediction accuracy;
  • PriceType – price type; default – close; This option does not affect the operation, if the parameter  NetFileName entered automatically generated name of the neural network file;
  • AddInvertedPrice –  on / off mode, when the set of samples are added to the inverted price movements; default – true; this means that the number of samples will be doubled;
  • NetFileName – filename trained neural network, the generated script SOMFX1Builder; if the name is formed automatically, it includes a number of necessary components to restore the above parameters; so the user can fill in only this one parameter; Filename following structure: SOM-V-D-SYMBOL-TF-YYYYMMDDHHMM-YYYYMMDDHHMM-P.candlemap, where V – PatternSize, D – GridSize, SYMBOL – the current symbol, TF – the current timeframe, YYYYMMDDHHMM – LearnStart and LearnStop respectively; P – PriceType;
  • CellSize – cell size when map visualization on the graph (if ShowMap equals true); Default – 20, which is suitable for up to 10 cards; to baboutlshih cards to choose the cell size smaller, or two of the map will be blocked;
  • PrintData – enable / disable the output of debugging messages in the log; default – false;
  • ShowMap – map display option on the chart in the same way as is done in the indicator SOMFX1; default – false;

If some parameters are incorrect, the indicator displays an error message in the log. For example, the card can only be loaded on the same character and the same timeframe, in which she studied.

SOMFX1Predictor

Related posts

  • ST Main Pivots

    ST Main Pivots ST Main Pivots – indicator that seeks reversal points at predetermined intervals (days, weeks, months, years, etc.). Then, all extremes…

  • Multi Timeframe Indicator

    Multi Timeframe Indicator Multi Timeframe Indicator combines many functions and can be considered as a universal trader’s assistant in the…

  • MTF Stochastic

    MTF Stochastic Indicator displays the graph data-Stochastic oscillator older time interval. The main and the signal lines are displayed in an additional…

Next posts

Leave a Reply

Your email address will not be published. Required fields are marked *