Machine Learning Moving Average [LuxAlgo]The Machine Learning Moving Average (MLMA) is a responsive moving average making use of the weighting function obtained Gaussian Process Regression method. Characteristic such as responsiveness and smoothness can be adjusted by the user from the settings.
The moving average also includes bands, used to highlight possible reversals.
🔶 USAGE
The Machine Learning Moving Average smooths out noisy variations from the price, directly estimating the underlying trend in the price.
A higher "Window" setting will return a longer-term moving average while increasing the "Forecast" setting will affect the responsiveness and smoothness of the moving average, with higher positive values returning a more responsive moving average and negative values returning a smoother but less responsive moving average.
Do note that an excessively high "Forecast" setting will result in overshoots, with the moving average having a poor fit with the price.
The moving average color is determined according to the estimated trend direction based on the bands described below, shifting to blue (default) in an uptrend and fushia (default) in downtrends.
The upper and lower extremities represent the range within which price movements likely fluctuate.
Signals are generated when the price crosses above or below the band extremities, with turning points being highlighted by colored circles on the chart.
🔶 SETTINGS
Window: Calculation period of the moving average. Higher values yield a smoother average, emphasizing long-term trends and filtering out short-term fluctuations.
Forecast: Sets the projection horizon for Gaussian Process Regression. Higher values create a more responsive moving average but will result in more overshoots, potentially worsening the fit with the price. Negative values will result in a smoother moving average.
Sigma: Controls the standard deviation of the Gaussian kernel, influencing weight distribution. Higher Sigma values return a longer-term moving average.
Multiplicative Factor: Adjusts the upper and lower extremity bounds, with higher values widening the bands and lowering the amount of returned turning points.
🔶 RELATED SCRIPTS
Machine-Learning-Gaussian-Process-Regression
SuperTrend-AI-Clustering
Ml
Machine Learning RSI [BackQuant]Machine Learning RSI
The Machine Learning RSI is a cutting-edge trading indicator that combines the power of Relative Strength Index (RSI) with Machine Learning (ML) clustering techniques to dynamically determine overbought and oversold thresholds. This advanced indicator adapts to market conditions in real-time, offering traders a robust tool for identifying optimal entry and exit points with increased precision.
Core Concept: Relative Strength Index (RSI)
The RSI is a well-known momentum oscillator that measures the speed and change of price movements, oscillating between 0 and 100. Typically, RSI values above 70 are considered overbought, and values below 30 are considered oversold. However, static thresholds may not be effective in all market conditions.
This script enhances the RSI by integrating a dynamic thresholding system powered by Machine Learning clustering, allowing it to adapt thresholds based on historical RSI behavior and market context.
Machine Learning Clustering for Dynamic Thresholds
The Machine Learning (ML) component uses clustering to calculate dynamic thresholds for overbought and oversold levels. Instead of relying on fixed RSI levels, this indicator clusters historical RSI values into three groups using a percentile-based initialization and iterative optimization:
Cluster 1: Represents lower RSI values (typically associated with oversold conditions).
Cluster 2: Represents mid-range RSI values.
Cluster 3: Represents higher RSI values (typically associated with overbought conditions).
Dynamic thresholds are determined as follows:
Long Threshold: The upper centroid value of Cluster 3.
Short Threshold: The lower centroid value of Cluster 1.
This approach ensures that the indicator adapts to the current market regime, providing more accurate signals in volatile or trending conditions.
Smoothing Options for RSI
To further enhance the effectiveness of the RSI, this script allows traders to apply various smoothing methods to the RSI calculation, including:
Simple Moving Average (SMA)
Exponential Moving Average (EMA)
Weighted Moving Average (WMA)
Hull Moving Average (HMA)
Linear Regression (LINREG)
Double Exponential Moving Average (DEMA)
Triple Exponential Moving Average (TEMA)
Adaptive Linear Moving Average (ALMA)
T3 Moving Average
Traders can select their preferred smoothing method and adjust the smoothing period to suit their trading style and market conditions. The option to smooth the RSI reduces noise and makes the indicator more reliable for detecting trends and reversals.
Long and Short Signals
The indicator generates long and short signals based on the relationship between the RSI value and the dynamic thresholds:
Long Signals: Triggered when the RSI crosses above the long threshold, signaling bullish momentum.
Short Signals: Triggered when the RSI falls below the short threshold, signaling bearish momentum.
These signals are dynamically adjusted to reflect real-time market conditions, making them more robust than static RSI signals.
Visualization and Clustering Insights
The Machine Learning RSI provides an intuitive and visually rich interface, including:
RSI Line: Plotted in real-time, color-coded based on its position relative to the dynamic thresholds (green for long, red for short, gray for neutral).
Dynamic Threshold Lines: The script plots the long and short thresholds calculated by the ML clustering process, providing a clear visual reference for overbought and oversold levels.
Cluster Plots: Each RSI cluster is displayed with distinct colors (green, orange, and red) to give traders insights into how RSI values are grouped and how the dynamic thresholds are derived.
Customization Options
The Machine Learning RSI is highly customizable, allowing traders to tailor the indicator to their preferences:
RSI Settings : Adjust the RSI length, source price, and smoothing method to match your trading strategy.
Threshold Settings : Define the range and step size for clustering thresholds, allowing you to fine-tune the clustering process.
Optimization Settings : Control the performance memory, maximum clustering steps, and maximum data points for ML calculations to ensure optimal performance.
UI Settings : Customize the appearance of the RSI plot, dynamic thresholds, and cluster plots. Traders can also enable or disable candle coloring based on trend direction.
Alerts and Automation
To assist traders in staying on top of market movements, the script includes alert conditions for key events:
Long Signal: When the RSI crosses above the long threshold.
Short Signal: When the RSI crosses below the short threshold.
These alerts can be configured to notify traders in real-time, enabling timely decisions without constant chart monitoring.
Trading Applications
The Machine Learning RSI is versatile and can be applied to various trading strategies, including:
Trend Following: By dynamically adjusting thresholds, this indicator is effective in identifying and following trends in real-time.
Reversal Trading: The ML clustering process helps identify extreme RSI levels, offering reliable signals for reversals.
Range-Bound Trading: The dynamic thresholds adapt to market conditions, making the indicator suitable for trading in sideways markets where static thresholds often fail.
Final Thoughts
The Machine Learning RSI represents a significant advancement in RSI-based trading indicators. By integrating Machine Learning clustering techniques, this script overcomes the limitations of static thresholds, providing dynamic, adaptive signals that respond to market conditions in real-time. With its robust visualization, customizable settings, and alert capabilities, this indicator is a powerful tool for traders seeking to enhance their momentum analysis and improve decision-making.
As always, thorough backtesting and integration into a broader trading strategy are recommended to maximize the effectiveness!
ANN Trend PredictionThis trend indicator utilizes an artificial neural network (ANN) to predict the next market reversal within a certain range of previous candles. The larger the range of previous candles you set, the fewer reversals will be predicted, and trends will tend to last longer.
The ANN is trained on the BTCUSD 4-hour chart, so using it on other assets or timeframes may yield suboptimal results. It takes three input values: the closing price, the Stochastic RSI, and a Choppiness Indicator. Based on these inputs, the ANN categorizes the current candle as part of an uptrend, downtrend, or as undefined.
Compared to an EMA-based trend indicator, this ANN identifies reversals several candles earlier. It achieves this by detecting subtle patterns in the input values that typically appear before a market turnaround. These patterns are somewhat specific to that chosen asset and timeframe.
The results are displayed using rows of triangles that indicate the predicted price direction. The price levels of the triangles correspond to the closing price at the last reversal. The area between the triangle row and the price is colored green if the ANN correctly predicted the move, and red if it did not.
This indicator is designed to showcase the capabilities and potential of ANNs, and is not intended for actual trading use. The ANN can be trained on any other input values, assets and timeframes for several predictions tasks.
You can use the Predicted_Trend_Signal of this Indicator in any backtest indicator. In the Backtester just grap the Predicted_Trend_Signal. downtrend = 1, uptrend = -1, undefined = 0
Feel free to write me a comment.
Machine Learning Cross-Validation Split & Batch HighlighterThis indicator is designed for traders and analysts who employ Machine Learning (ML) techniques for cross-validation in financial markets.
The script visually segments a selected range of historical price data into splits and batches, helping in the assessment of model performance over different market conditions.
User
Theory
In ML, cross-validation is a technique to assess the generalizability of a model, typically by partitioning the data into a set of "folds" or "splits." Each split acts as a validation set, while the others form the training set. This script takes a unique approach by considering the sequential nature of financial time series data, where random shuffling of data (as in traditional cross-validation) can disrupt the temporal order, leading to misleading results.
Chronological Integrity of Splits
Even if the order of the splits is shuffled for cross-validation purposes, the data within each split remains in its original chronological sequence. This feature is crucial for time series analysis, as it respects the inherent order-dependency of financial markets. Thus, each split can be considered a microcosm of market behavior, maintaining the integrity of trends, cycles, and patterns that could be disrupted by random sampling.
The script allows users to define the number of splits and the size of each batch within a split. By doing so, it maintains the chronological sequence of the data, ensuring that the validation set is representative of a future time period that the model would predict.
www.tradingview.com
Parameters
Number of Splits: Defines how many segments the selected data range will be divided into. Each split serves as a standalone testing ground for the ML model. (Up to 24)
Batch Size: Determines the number of bars (candles) in each batch within a split. Smaller batches can help pinpoint overfitting at a finer granularity.
Start Index: The bar index from where the historical data range begins. It sets the starting point for data analysis.
End Index: The bar index where the historical data range ends. It marks the cutoff for data to be included in the model assessment.
Usage
To use this script effectively:
1 - Input the Start Index and End Index to define the historical data range you wish to analyze.
2 - Adjust the Number of Splits to create multiple validation sets for cross-validation.
3 - Set the Batch Size to control the granularity of each validation set within the splits.
4 - The script will highlight the background of each batch within the splits using alternating shades, allowing for a clear visual distinction of the data segmentation.
By maintaining the temporal sequence and allowing for adjustable granularity, the "ML Split and Batch Highlighter" aids in creating a robust validation framework for time series forecasting models in finance.
ML - Momentum Index (Pivots)Building upon the innovative foundations laid by Zeiierman's Machine Learning Momentum Index (MLMI), this variation introduces a series of refinements and new features aimed at bolstering the model's predictive accuracy and responsiveness. Licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License (CC BY-NC-SA 4.0), my adaptation seeks to enhance the original by offering a more nuanced approach to momentum-based trading.
Key Features :
Pivot-Based Analysis: Shifting focus from trend crosses to pivot points, this version employs pivot bars to offer a distinct perspective on market momentum, aiding in the identification of critical reversal points.
Extended Parameter Set: By integrating additional parameters for making predictions, the model gains improved adaptability, allowing for finer tuning to match market conditions.
Dataset Size Limitation: To ensure efficiency and mitigate the risk of calculation timeouts, a cap on the dataset size has been implemented, balancing between comprehensive historical analysis and computational agility.
Enhanced Price Source Flexibility: Users can select between closing prices or (suggested) OHLC4 as the basis for calculations, tailoring the indicator to different analysis preferences and strategies.
This adaptation not only inherits the robust framework of the original MLMI but also introduces innovations to enhance its utility in diverse trading scenarios. Whether you're looking to refine your short-term trading tactics or seeking stable indicators for long-term strategies, the ML - Momentum Index (Pivots) offers a versatile tool to navigate the complexities of the market.
For a deeper understanding of the modifications and to leverage the full potential of this indicator, users are encouraged to explore the tooltips and documentation provided within the script.
The Momentum Indicator calculations have been transitioned to the MLMomentumIndex library, simplifying the process of integration. Users can now seamlessly incorporate the momentumIndexPivots function into their scripts to conduct detailed momentum analysis with ease.
Machine Learning using Neural Networks | EducationalThe script provided is a comprehensive illustration of how to implement and execute a simplistic Neural Network (NN) on TradingView using PineScript.
It encompasses the entire workflow from data input, weight initialization, implicit neuron calculation, feedforward computation, backpropagation for weight adjustments, generating predictions, to visualizing the Mean Squared Error (MSE) Loss Curve for monitoring the training phase.
In the visual example above, you can see that the prediction is not aligned with the actual value. This is intentional for demonstrative purposes, and by incrementing the Epochs or Learning Rate, you will see these two values converge as the accuracy increases.
Hyperparameters:
Learning Rate, Epochs, and the choice between Simple Backpropagation and a verbose version are declared as script inputs, allowing users to tailor the training process.
Initialization:
Random initialization of weight matrices (w1, w2) is performed to ensure asymmetry, promoting effective gradient updates. A seed is added for reproducibility.
Utility Functions:
Functions for matrix randomization, sigmoid activation, MSE loss calculation, data normalization, and standardization are defined to streamline the computation process.
Neural Network Computation:
The feedforward function computes the hidden and output layer values given the input.
Two variants of the backpropagation function are provided for weight adjustment, with one offering a more verbose step-by-step computation of gradients.
A wrapper train_nn function iterates through epochs, performing feedforward, loss computation, and backpropagation in each epoch while logging and collecting loss values.
Training Invocation:
The input data is prepared by normalizing it to a value between 0 and 1 using the maximum standardized value, and the training process is invoked only on the last confirmed bar to preserve computational resources.
Output Forecasting and Visualization:
Post training, the NN's output (predicted price) is computed, standardized and visualized alongside the actual price on the chart.
The MSE loss between the predicted and actual prices is visualized, providing insight into the prediction accuracy.
Optionally, the MSE Loss Curve is plotted on the chart, illustrating the loss trajectory through epochs, assisting in understanding the training performance.
Customizable Visualization:
Various inputs control visualization aspects like Chart Scaling, Chart Horizontal Offset, and Chart Vertical Offset, allowing users to adapt the visualization to their preference.
-------------------------------------------------------
The following is this Neural Network structure, consisting of one hidden layer, with two hidden neurons.
Through understanding the steps outlined in my code, one should be able to scale the NN in any way they like, such as changing the input / output data and layers to fit their strategy ideas.
Additionally, one could forgo the backpropagation function, and load their own trained weights into the w1 and w2 matrices, to have this code run purely for inference.
-------------------------------------------------------
While this demonstration does create a “prediction”, it is on historical data. The purpose here is educational, rather than providing a ready tool for non-programmer consumers.
Normally in Machine Learning projects, the training process would be split into two segments, the Training and the Validation parts. For the purpose of conveying the core concept in a concise and non-repetitive way, I have foregone the Validation part. However, it is merely the application of your trained network on new data (feedforward), and monitoring the loss curve.
Essentially, checking the accuracy on “unseen” data, while training it on “seen” data.
-------------------------------------------------------
I hope that this code will help developers create interesting machine learning applications within the Tradingview ecosystem.
Relational Quadratic Kernel Channel [Vin]The Relational Quadratic Kernel Channel (RQK-Channel-V) is designed to provide more valuable potential price extremes or continuation points in the price trend.
Example:
Usage:
Lookback Window: Adjust the "Lookback Window" parameter to control the number of previous bars considered when calculating the Rational Quadratic Estimate. Longer windows capture longer-term trends, while shorter windows respond more quickly to price changes.
Relative Weight: The "Relative Weight" parameter allows you to control the importance of each data point in the calculation. Higher values emphasize recent data, while lower values give more weight to historical data.
Source: Choose the data source (e.g., close price) that you want to use for the kernel estimate.
ATR Length: Set the length of the Average True Range (ATR) used for channel width calculation. A longer ATR length results in wider channels, while a shorter length leads to narrower channels.
Channel Multipliers: Adjust the "Channel Multiplier" parameters to control the width of the channels. Higher multipliers result in wider channels, while lower multipliers produce narrower channels. The indicator provides three sets of channels, each with its own multiplier for flexibility.
Details:
Rational Quadratic Kernel Function:
The Rational Quadratic Kernel Function is a type of smoothing function used to estimate a continuous curve or line from discrete data points. It is often used in time series analysis to reduce noise and emphasize trends or patterns in the data.
The formula for the Rational Quadratic Kernel Function is generally defined as:
K(x) = (1 + (x^2) / (2 * α * β))^(-α)
Where:
x represents the distance or difference between data points.
α and β are parameters that control the shape of the kernel. These parameters can be adjusted to control the smoothness or flexibility of the kernel function.
In the context of this indicator, the Rational Quadratic Kernel Function is applied to a specified source (e.g., close prices) over a defined lookback window. It calculates a smoothed estimate of the source data, which is then used to determine the central value of the channels. The kernel function allows the indicator to adapt to different market conditions and reduce noise in the data.
The specific parameters (length and relativeWeight) in your indicator allows to fine-tune how the Rational Quadratic Kernel Function is applied, providing flexibility in capturing both short-term and long-term trends in the data.
To know more about unsupervised ML implementations, I highly recommend to follow the users, @jdehorty and @LuxAlgo
Optimizing the parameters:
Lookback Window (length): The lookback window determines how many previous bars are considered when calculating the kernel estimate.
For shorter-term trading strategies, you may want to use a shorter lookback window (e.g., 5-10).
For longer-term trading or investing, consider a longer lookback window (e.g., 20-50).
Relative Weight (relativeWeight): This parameter controls the importance of each data point in the calculation.
A higher relative weight (e.g., 2 or 3) emphasizes recent data, which can be suitable for trend-following strategies.
A lower relative weight (e.g., 1) gives more equal importance to historical and recent data, which may be useful for strategies that aim to capture both short-term and long-term trends.
ATR Length (atrLength): The length of the Average True Range (ATR) affects the width of the channels.
Longer ATR lengths result in wider channels, which may be suitable for capturing broader price movements.
Shorter ATR lengths result in narrower channels, which can be helpful for identifying smaller price swings.
Channel Multipliers (channelMultiplier1, channelMultiplier2, channelMultiplier3): These parameters determine the width of the channels relative to the ATR.
Adjust these multipliers based on your risk tolerance and desired channel width.
Higher multipliers result in wider channels, which may lead to fewer signals but potentially larger price movements.
Lower multipliers create narrower channels, which can result in more frequent signals but potentially smaller price movements.
Fusion: Machine Learning SuiteThe Fusion: Machine Learning Suite combines multiple technical analysis dimensions and harnesses the predictive power of machine learning, seamlessly integrating a diverse array of classic and novel indicators to deliver precision, adaptability, and innovation.
Features and Capabilities
Multidimensional Analysis: Fusion: MLS integrates various technical analysis dimensions to offer a more comprehensive perspective.
Machine Learning Integration: Utilizing ML algorithms, Fusion: MLS offers adaptability to market changes.
Custom Indicators: Including dimensions like "Moon Lander", "Cap Line" and "Z-Pack" the indicator expands the scope of traditional technical analysis methods.
Tailored Customization: With customization options, Fusion: MLS allows traders to configure the tool to suit their specific strategies and market focus.
In the following sections, we'll explore the features and settings of Fusion: MLS in detail, providing insights into how it can be utilized.
Major Features and Settings
The indicator consists of several core components and settings, each designed to provide specific functionalities and insights. Here's an in-depth look:
Machine Learning Component
Distance Classifier: A Strategic Approach to Market Analysis
In the world of trading and investment, the ability to classify and predict price movements is paramount. Machine learning offers powerful tools for this purpose.
The Fusion: MLS indicator among others incorporates an Approximate Nearest Neighbors (ANN)* algorithm, a machine learning classification technique, and allows the selection of various distance functions .
This flexibility sets Fusion: MLS apart from existing solutions. The available distance functions include:
Euclidean: Standard distance metric, commonly used as a default.
Chebyshev: Also known as maximum value distance.
Manhattan: Sum of absolute differences.
Minkowski: Generalized metric that includes Euclidean and Manhattan as special cases.
Mahalanobis: Measures distance between points in a correlated space.
Lorentzian: Known for its robustness to outliers and noise.
*For a deeper understanding of the Approximate Nearest Neighbors (ANN) algorithm, traders are encouraged to refer to the relevant articles that can be found in the public domain.
Alternative scoring system
Fusion: MLS also includes a custom scoring alternative based on directional price action.
"Combined: Directional" and "Alpha: Directional" scoring types represent our own directional change algorithm, simple yet effective in displaying trend direction changes early on. They are visualized by color changes when scoring becomes below or above zero.
Changes in scoring quickly reflect shifts in buyer and seller sentiment.
Traders may choose signals by Color Change in the indicator settings to get alerts when scoring color shifts, not waiting until the histogram crosses the zero level.
Application in Trading
Machine learning classification has become an integral part of modern trading, offering innovative ways to analyze and interpret financial data.
Many algorithmic trading systems leverage ML classification to automate trading decisions. By continuously learning from real-time data, these systems can adapt to changing market conditions and execute trades with increased efficiency and accuracy.
ML classification allows for the development of tailored trading strategies as traders can select specific algorithms, dimensions, and filters that align with their trading style, goals, and the particular market they are operating.
We have integrated ML classification with traditional trading tools, such as moving averages and technical indicators. This fusion creates a more robust analysis framework, combining the strengths of classical techniques with the adaptability of machine learning.
Whether used independently or in conjunction with other tools, ML classification represents a significant advancement in trading technology, opening new avenues for exploration, innovation, and success in the financial world.
ML: Weighting System
The Fusion: MLS indicator introduces a unique weighting system that allows traders to customize the influence of various technical indicators in the machine learning process. This feature is not only innovative but also provides a level of control and adaptability that sets it apart from other indicators.
Customizable Weights
The weighting system allows users to assign specific weights to different indicators such as Moon Lander, RSI, MACD, Money Flow, Bollinger Bands, Cap Line, Z-Pack, Squeeze Momentum*, and MA Crossover. These weights can be adjusted manually, providing the ability to emphasize or de-emphasize specific indicators based on the trader's strategy or market conditions.
*Note, we determined via testing that the popular "Squeeze" indicator can actually be well replicated by simply using inputs of 15 & 199 in the bedrock indicator - MACD ; while we employed the standard "Squeeze" formula (developed by J. Carter ) in Fusion: MLS, traders are hereby made aware of our research findings regarding such.
The weighting system's importance lies in its ability to provide a more nuanced and personalized analysis. By adjusting the weights of different indicators a trader focusing on momentum strategies might assign higher weights to the Squeeze Momentum and MA Crossover indicators, while a trader looking for volatility might emphasize RSI and Bollinger Bands.
The ability to customize weights adds a layer of complexity and adaptability that is rare in standard machine-learning indicators.
Custom Indicators: Moon Lander
The "Moon Lander" is not just a catchy name; it's a robust feature inspired by principles from aerospace engineering and offers a unique perspective on trading analysis. Here's a conceptual overview:
Fast EMA and Kalman Matrix
"Moon Lander" incorporates both a Fast Exponential Moving Average (EMA) and a Kalman Matrix in its design. These two elements are combined to create a histogram, providing a specific approach to data analysis.
The Kalman Matrix, or Kalman Filter, is a mathematical concept used for estimating variables that can be measured indirectly and contain noise or uncertainty. It's a standard tool in machine learning and control systems, known for its ability to provide optimal estimates based on observed data.
Kalman Filter: A Navigational Tool
The Kalman filter, an essential part of "Moon Lander," is a mathematical concept known for its applications in navigation and control systems used by NASA in the apollo program :
Guidance in Uncertainty: Just as the Kalman filter helped guide complex aerospace missions through uncertain paths, it assists traders in navigating the often unpredictable financial markets.
Filtering Noise: In trading, the Kalman filter serves to filter out market noise, allowing traders to focus on the underlying trends.
Predictive Capabilities: Its ability to predict future states makes it a valuable tool for forecasting market movements and trend directions.
Custom Indicators: Cap Line and Z-Pack
Fusion: MLS integrates our additional proprietary custom indicators that have been published on TradingView earlier:
Cap Line: Delve into the specific functionalities and applications of our proprietary "Cap Line" indicator in the published description on TradingView.
Z-Pack: Explore the analytical perspectives, focused on the z-score methodology, and custom "Z-Pack" indicator by reviewing the published description on TradingView.
Buy/Sell Signal Generation Algorithms
Fusion: MLS offers various options for generating buy/sell signals, tailored to different trading strategies and perspectives:
Fusion: Allows traders to select any number of dimensions to receive buy/sell signals from, offering customized signal generation.
ML: Utilizes the machine learning ANN distance for signal generation.
Color Change: Generates signals by selected scoring type color change.
Displayed Dimension, Alpha Dimension: Generate signals based on specific selected dimensions.
These algorithms provide flexibility in determining buy/sell signals, catering to different trading styles and market conditions.
Filters
Filters are used to refine and selectively include or exclude signals based on specific criteria. Rather than generating signals, these filters act as gatekeepers, ensuring that only the signals meeting certain conditions are considered. Here's an overview of the filters used:
Dynamic State Predictor (DSP)
The DSP employs the Kalman Matrix to evaluate existing signals by comparing the fast and slow-moving averages, both processed through the Kalman Matrix. Based on the relationship between these averages, the DSP may exclude specific signals, depending on whether they align with upward or downward trends.
Average Directional Index (ADX)
The ADX filter evaluates the strength of existing trends and filters out signals that do not meet the specified ADX threshold and length, focusing on significant market movements.
Feature Engineering: RSI
Applies a filter to the existing signals, clearing out those that do not meet the criteria for RSI overbought or oversold threshold condition.
Feature Engineering: MACD
Assesses existing signals to identify changes in the strength, direction, momentum, and duration of a trend, filtering out those that do not align with MACD trend direction.
The Visual Component
The machine learning component is an internal component. However, the indicator also offers an equally important and useful visual component. It is a graphical representation of the multiple technical analysis dimensions, that can be combined in various ways (where the name "Fusion" comes from), allowing traders to visualize the underlying data and its analysis.
Displayed Dimension: Visualization and Normalization
The Fusion: MLS indicator offers a "Displayed Dimension" feature that visualizes various dimensions as a histogram. These dimensions may include RSI, MAs, BBs, MACD, etc.
RSI Dimension on the image + ML signals
Normalization: Each dimension is normalized. If any dimension has extreme values, a Fisher transformation is applied to bring them within a reasonable range.
Combined Dimension: When selecting the "Combined" option , the normalized values of the selected dimensions are combined using techniques such as standardization, normalization, or winsorization. This flexibility enables tailored visualization and analysis.
Alpha Dimension: Enhancing Analysis
The "Alpha Dimension" feature allows traders to select an additional dimension alongside the Displayed Dimension. This facilitates a combined analysis, enhancing the depth of insights.
Theme Selection
Fusion: MLS offers various themes such as "Sailfish", "Iceberg", "Moon", "Perl", "Candy" and "Monochrome" Traders can select a theme that resonates with their preference, enhancing visual appeal. There is also a "Custom" theme available that allows the user to choose the colors of the theme.
Customizing Fusion: MLS for Various Markets and Strategies
Fusion: MLS is designed with customization in mind. Traders can tailor the indicator to suit various markets and trading strategies. Selecting specific dimensions allows it to align with individual trading goals.
Selecting Dimensions: Choose the dimensions that resonate with your trading approach, whether focusing on trend-following, momentum, or other strategies.
Adjusting Parameters: Fine-tune the parameters of each dimension, including custom ones like "Moon Lander," to suit specific market conditions.
Theme Customization: Select a theme that aligns with your visual preferences, enhancing your chart's readability and appeal.
Utilizing Research: Leverage the underlying algorithms and research, such as machine learning classification by ANN and the Kalman filter, to deepen your understanding and application of Fusion: MLS.
Alerts
The indicator includes an alerting system that notifies traders when new buy or sell signals are detected.
Disclaimer
The information provided herein is intended for informational purposes only and should not be construed as investment advice, endorsement, nor a recommendation to buy or sell any financial instruments. Fusion: MLS is a technical analysis tool, and like all tools, it should be used with caution and in conjunction with other forms of analysis.
Traders and investors are encouraged to consult with a licensed financial professional and conduct their own research before making any trading or investment decisions. Past performance of the Fusion: MLS indicator or any trading strategy does not guarantee future results, and all trading involves risk. Users of Fusion: MLS should understand the underlying algorithms and assumptions and consider their individual risk tolerance and investment goals when using this tool.
Machine Learning Regression Trend [LuxAlgo]The Machine Learning Regression Trend tool uses random sample consensus (RANSAC) to fit and extrapolate a linear model by discarding potential outliers, resulting in a more robust fit.
🔶 USAGE
The proposed tool can be used like a regular linear regression, providing support/resistance as well as forecasting an estimated underlying trend.
Using RANSAC allows filtering out outliers from the input data of our final fit, by outliers we are referring to values deviating from the underlying trend whose influence on a fitted model is undesired. For financial prices and under the assumptions of segmented linear trends, these outliers can be caused by volatile moves and/or periodic variations within an underlying trend.
Adjusting the "Allowed Error" numerical setting will determine how sensitive the model is to outliers, with higher values returning a more sensitive model. The blue margin displayed shows the allowed error area.
The number of outliers in the calculation window (represented by red dots) can also be indicative of the amount of noise added to an underlying linear trend in the price, with more outliers suggesting more noise.
Compared to a regular linear regression which does not discriminate against any point in the calculation window, we see that the model using RANSAC is more conservative, giving more importance to detecting a higher number of inliners.
🔶 DETAILS
RANSAC is a general approach to fitting more robust models in the presence of outliers in a dataset and as such does not limit itself to a linear regression model.
This iterative approach can be summarized as follow for the case of our script:
Step 1: Obtain a subset of our dataset by randomly selecting 2 unique samples
Step 2: Fit a linear regression to our subset
Step 3: Get the error between the value within our dataset and the fitted model at time t , if the absolute error is lower than our tolerance threshold then that value is an inlier
Step 4: If the amount of detected inliers is greater than a user-set amount save the model
Repeat steps 1 to 4 until the set number of iterations is reached and use the model that maximizes the number of inliers
🔶 SETTINGS
Length: Calculation window of the linear regression.
Width: Linear regression channel width.
Source: Input data for the linear regression calculation.
🔹 RANSAC
Minimum Inliers: Minimum number of inliers required to return an appropriate model.
Allowed Error: Determine the tolerance threshold used to detect potential inliers. "Auto" will automatically determine the tolerance threshold and will allow the user to multiply it through the numerical input setting at the side. "Fixed" will use the user-set value as the tolerance threshold.
Maximum Iterations Steps: Maximum number of allowed iterations.
Universal Moving Average Convergence DivergenceI changed MACD formula to divergence of (MA26/MA12 - 1).
And its make it more useful.
Cuz:
1) comparability with all other coins with different prices.
2) fix small numbers in low price coines like shiba
3) making a good indicator like RSI to use it for optimization and ML/AI projects as a variable
Most important thing about this indicator is that its Universal
Now you can compare the UMACD of Shiba with Bitcoin without any problem in matamatics space.No need to use virtuality and its important in Optimization problems that we rediuse the problem from a picture to a number(A plot to a list of numbers)
If we don't care about exagrated pumps and dumps, we can say to it Normalized-MACD too. Cuz in normal situations its MAX ≈ 0.1 and MIN ≈ -0.1
WIPNNetworkLibrary "WIPNNetwork"
this is a work in progress (WIP) and prone to have some errors, so use at your own risk...
let me know if you find any issues..
Method for a generalized Neural Network.
network(x) Generalized Neural Network Method.
Parameters:
x : TODO: add parameter x description here
Returns: TODO: add what function returns
FunctionNNLayerLibrary "FunctionNNLayer"
Generalized Neural Network Layer method.
function(inputs, weights, n_nodes, activation_function, bias, alpha, scale) Generalized Layer.
Parameters:
inputs : float array, input values.
weights : float array, weight values.
n_nodes : int, number of nodes in layer.
activation_function : string, default='sigmoid', name of the activation function used.
bias : float, default=1.0, bias to pass into activation function.
alpha : float, default=na, if required to pass into activation function.
scale : float, default=na, if required to pass into activation function.
Returns: float
FunctionNNPerceptronLibrary "FunctionNNPerceptron"
Perceptron Function for Neural networks.
function(inputs, weights, bias, activation_function, alpha, scale) generalized perceptron node for Neural Networks.
Parameters:
inputs : float array, the inputs of the perceptron.
weights : float array, the weights for inputs.
bias : float, default=1.0, the default bias of the perceptron.
activation_function : string, default='sigmoid', activation function applied to the output.
alpha : float, default=na, if required for activation.
scale : float, default=na, if required for activation.
@outputs float
MLActivationFunctionsLibrary "MLActivationFunctions"
Activation functions for Neural networks.
binary_step(value) Basic threshold output classifier to activate/deactivate neuron.
Parameters:
value : float, value to process.
Returns: float
linear(value) Input is the same as output.
Parameters:
value : float, value to process.
Returns: float
sigmoid(value) Sigmoid or logistic function.
Parameters:
value : float, value to process.
Returns: float
sigmoid_derivative(value) Derivative of sigmoid function.
Parameters:
value : float, value to process.
Returns: float
tanh(value) Hyperbolic tangent function.
Parameters:
value : float, value to process.
Returns: float
tanh_derivative(value) Hyperbolic tangent function derivative.
Parameters:
value : float, value to process.
Returns: float
relu(value) Rectified linear unit (RELU) function.
Parameters:
value : float, value to process.
Returns: float
relu_derivative(value) RELU function derivative.
Parameters:
value : float, value to process.
Returns: float
leaky_relu(value) Leaky RELU function.
Parameters:
value : float, value to process.
Returns: float
leaky_relu_derivative(value) Leaky RELU function derivative.
Parameters:
value : float, value to process.
Returns: float
relu6(value) RELU-6 function.
Parameters:
value : float, value to process.
Returns: float
softmax(value) Softmax function.
Parameters:
value : float array, values to process.
Returns: float
softplus(value) Softplus function.
Parameters:
value : float, value to process.
Returns: float
softsign(value) Softsign function.
Parameters:
value : float, value to process.
Returns: float
elu(value, alpha) Exponential Linear Unit (ELU) function.
Parameters:
value : float, value to process.
alpha : float, default=1.0, predefined constant, controls the value to which an ELU saturates for negative net inputs. .
Returns: float
selu(value, alpha, scale) Scaled Exponential Linear Unit (SELU) function.
Parameters:
value : float, value to process.
alpha : float, default=1.67326324, predefined constant, controls the value to which an SELU saturates for negative net inputs. .
scale : float, default=1.05070098, predefined constant.
Returns: float
exponential(value) Pointer to math.exp() function.
Parameters:
value : float, value to process.
Returns: float
function(name, value, alpha, scale) Activation function.
Parameters:
name : string, name of activation function.
value : float, value to process.
alpha : float, default=na, if required.
scale : float, default=na, if required.
Returns: float
derivative(name, value, alpha, scale) Derivative Activation function.
Parameters:
name : string, name of activation function.
value : float, value to process.
alpha : float, default=na, if required.
scale : float, default=na, if required.
Returns: float
MLLossFunctionsLibrary "MLLossFunctions"
Methods for Loss functions.
mse(expects, predicts) Mean Squared Error (MSE) " MSE = 1/N * sum ((y - y')^2) ".
Parameters:
expects : float array, expected values.
predicts : float array, prediction values.
Returns: float
binary_cross_entropy(expects, predicts) Binary Cross-Entropy Loss (log).
Parameters:
expects : float array, expected values.
predicts : float array, prediction values.
Returns: float
neutronix community bot ML + Alerts 4h-daily (mod. capissimo)Gm traders,
i have been a python programmer for some years studying artificial intelligence for general purpose; after some time i finally decided to have a look at some finance related stuff and scripts.
Moved by curiosity i've decided to make some but decisive modifications to a script i tried to use initially but without success: the LVQ machine learning strategy.
So after studying the charts and indicators, i have rewritten this script made by Capissimo and added heavy filtering thanks to vwap and vwma, then fixed repaint and other issues.
I hope you enjoy it and that it could increase your possibilities of success in trading.
HOW TO USE THE SCRIPT
Add the script to 3h+ charts like for example BTC 4h, 6h, 8h, 12h, daily. (In order for it to work on shorter timeframes charts you can try to change to lookback window but i dont advise it).
Change only rsi and volfilter(volume filtering) settings to try to find the best winrate. Leave dataset to open. Fyi the winrate isn't 100% accurate but can give you a raw vision of final results.
Use alerts included for trading and and in options click on 'Once per bar'. If you have checked 'Reverse Signals' in the control panel you have got more 'risky' signals so be advised if trading futures and stocks.
Exit trade signals not provided, so it is recommended the use of take profits and stop loss (1.5:1 ratio)
As always, the script is for study purposes. Do not risk more than you can spend!
Original LVQ-based strategy made by capissimo
Modified by gravisxv 13/10/2021
Minkowski Distance Factor Adaptive Period MACDHi, this script comes from the idea that Ricardo Santos' Minkovski Distance Function is transferred to the period as a factor.
Minkowski distance is used as a percentage factor with the help of Relative Strength Index function.
Minkowski Distance Function Script :
And thus an adaptive MACD was created.
This script can give much better results in more optimized larger periods.
I leave the decision to determine the periods and weights.
I used the weights of 9,12,26 and periods created with multiplied by factor.
Regards.
CBOE PCR Factor Dependent Variable Odd Generator This script is the my Dependent Variable Odd Generator script :
with the Put / Call Ratio ( PCR ) appended, only for CBOE and the instruments connected to it.
For CBOE this script is more accurate and faster than Dependent Variable Odd Generator. And the stagnant market odds are better and more realistic.
Do not use for timeframe periods less than 1 day.
Because PCR data may give repaint error.
My advice is to use the 1-week bars to gain insight into your analysis.
This code is open source under the MIT license. If you have any improvements or corrections to suggest, please send me a pull request via the github repository github.com
I hope it will help your work.Best regards!
