An Expert Advisor (EA) has been developed to automate the process of exporting the closing data of a specified number of candles, X, at regular intervals, every Z seconds. The primary function of this EA is to ensure seamless updates to a CSV file, which can then be readily accessed by Python scripts for further analysis and backtesting. This automation is designed to facilitate more efficient data handling and integration into trading strategies that require frequent data refreshes.

Read more...

The previously implemented EA feature allows for choosing a strategy option with either constant or variable position sizes. This supports result normalization according to the maximum drawdown and enables grouping strategies within specified limits. Manual selection of optimized strategy sets demonstrated effectiveness when combining them into multiple groups. However, scaling this approach can become labor-intensive.

The optimization process requires refining individual strategies with different criteria for each trading symbol and timeframe, including order types. From large parameter sets (20-50k), the top-performers for groups must be identified. Manually this is feasible only for a few levels, prompting the need for automation.

Automating group selection can start with testing the hypothesis that automation's quality meets or exceeds manual selection. By exporting optimization...

Read more...

In the realm of algorithmic trading, innovation is essential for staying competitive. A sophisticated Expert Advisor (EA) combines machine learning with traditional technical analysis for navigating forex markets. This EA employs an ONNX model alongside optimized technical indicators, such as SMA and EMA, to make informed trading decisions primarily for AAPL stock, but it is adaptable for other instruments.

Key features include dynamic lot sizing, trailing stops, and automatic market condition adjustments, creating a blend of modern technology and time-tested trading principles. The EA utilizes an ONNX model for price prediction, integrating trend-following techniques and adaptive parameter optimization. This multi-faceted approach considers both swift price changes and long-term trends.

The EA incorporates various technical indicators:

- **Simple Moving Average (SMA):** Uses an a...

Read more...

Mean Reversion strategy is effective on daily timeframes for major forex pairs. It is recommended to test on a demo account initially. To avoid trading with increasing lot sizes when losing a trade, set "IncreaseFactor=0".

Inputs:
- Use_TP_In_Money: Enable Take Profit in money (true/false)
- TP_In_Money: Take Profit value (10-100)
- Use_TP_In_Percent: Use Take Profit in percent (true/false)
- TP_In_Percent: Take Profit percentage (10-100)

Money Trailing Stop:
- Enable_Trailing: Enable using money trailing stop (true/false)
- TP_In_Money: Take Profit in current currency (25-200)
- Stop_Loss_In_Money: Stop Loss in current currency (1-20)

Additional Settings:
- Exit: Close trades if trend is adverse (true/false)
- BarsToCount: Number of bars to count (1-20)
- Lots: Lot size (0.01-1)
- Lots_Size_Exponent: Lot size exponent (1.01-2)
- IncreaseFactor: Increase lots from total margin o...

Read more...

Ridge regression is a method to estimate coefficients of multiple-regression models when independent variables are highly correlated. It offers improved efficiency in parameter estimation in exchange for tolerable bias. Ridge regression reduces high variance by introducing a small bias, providing a more reliable model in the long run.

Lasso (Least Absolute Shrinkage and Selection Operator) performs both variable selection and regularization to enhance prediction accuracy and interpretability. In situations where covariates are collinear, lasso can introduce soft thresholding to manage coefficients.

Bias is the inability to capture the true relationship between variables. Low bias implies fewer assumptions; high bias means more. Variance indicates the difference from expected values, with high variance indicating overfitting. Balancing bias and variance is essential for accurate pred...

Read more...

Attention developers and traders. This EA will terminate all active trades upon application. Ensure you are aware of your trading positions before deploying this Expert Advisor. Appropriate use of this tool can manage exposure and mitigate risk. Evaluate the implications for open positions in advance. Use this EA in environments where trade termination aligns with your broader strategy. For detailed implementation, review available documentation and consider testing in a controlled setting beforehand. Remember, strategic integration is key to leveraging automation effectively in your trading systems.

Read more...

The Gray Wolf Optimization (GWO) algorithm, developed in 2014, is a metaheuristic stochastic swarm intelligence algorithm inspired by the hunting behavior of gray wolf packs. The hierarchy among wolves includes alpha, beta, delta, and omega. Alpha wolves dominate, with betas and deltas aiding decision-making. Omegas, with the least influence, obey all higher-ranking wolves.

Mathematically, the alpha wolf represents the best solution, beta the second best, and delta the third. Omegas signify all other potential solutions. The algorithm's core stages, search, encircling, and attacking, help determine the optimal solution. During the search, alpha, beta, and delta wolves guide the pack toward the prey, while omegas continue to seek better solutions.

Encircling involves reassessing positions based on the fitness function, leading to refined prey location estimates. When attacking, wolv...

Read more...

A new Trade Management Script has been released for opening market orders efficiently. Users can define essential trade parameters:

- Order Type
- Stop Loss (in pips)
- Take Profit (in pips)
- Lot Size
- Slippage
- Magic Number for Trade
- Price for Placing Pending Orders

This script facilitates precise trade execution by allowing detailed input specifications. Utilize this tool to streamline your trading process and enhance order management. For visual reference, an example can be accessed via the provided image. Focus on optimizing your trading strategy with these configurable options.

Read more...

The complexity of financial markets is growing with technology. Over 95% of trading turnover is now generated by trading robots. Machine learning models are the logical next step, offering more robust trading systems than linear algorithms. Neural networks enable trading robots to analyze vast datasets, identify patterns, and forecast price movements effectively.

The development cycle of a trading robot includes stages such as data collection, processing, sample expansion, feature engineering, model selection and training, creating trading systems in Python, and monitoring trades. Python offers speed and flexibility in machine learning, with tools available to facilitate this process.

The project’s objective was to create a profitable machine learning model for trading using Python. The steps are:
- Collecting data from MetaTrader 5.
- Expanding the sample through data augmentation...

Read more...

The development of our Expert Advisor (EA) with a focus on economic data is progressing. Previously, we implemented a system to store economic data in a news calendar database and developed foundational classes for the expert's functionality. Today's update involves expanding these classes to facilitate trading based on economic data, aiming for future profitability.

Key updates include the addition of a new database view displaying unique events from the MQL5 economic calendar and new expert inputs for better filtering of economic data during trades, enhancing flexibility.

Notable UI Improvements:
- Concise, modern graphics with better responsiveness
- Display of terminal date and time highlights during news events
- Dynamic updates for news event details

Enhanced Filtering Options:
- New inputs for risk management, DST schedules, and news settings, providing granular control ove...

Read more...

In the previous segment, a Telegram-integrated Expert Advisor was linked with MQL5. Initial steps included setting up the application and developing code to send and receive messages via Telegram. This foundational bot setup enabled straightforward MQL5 messaging to Telegram.

Proceeding to the next step, the focus shifts to transmitting trading signals using MQL5. This enhanced Expert Advisor opens and closes trades based on preset conditions while sending signals to a Telegram group chat. The trading signals have been refined for clarity and conciseness, ensuring nearly real-time updates.

We generate signals using the moving average crossover system. A simple method involving shorter-term and longer-term moving averages identifies potential buy or sell opportunities. Upon detecting a crossover, signals are formatted and sent to Telegram, including the asset name, crossover directio...

Read more...

A new script has been introduced aimed at streamlining order management by closing all open orders simultaneously, excluding pending orders. This script is designed to enhance operational efficiency by allowing traders to execute a bulk closure of currently active trades.

By specifically excluding pending orders, it ensures that future strategies remain unaffected while addressing immediate risk management needs. This tool is especially useful for minimizing manual intervention during volatile market conditions.

Adopting this script can significantly reduce the time spent on individual order closures, thus providing a more efficient workflow. This method integrates seamlessly with existing trading infrastructure, making it a valuable addition to any trader's toolkit.

Read more...

In previous discussions, we analyzed an S&P 500 trading strategy employing high-weight stocks in the index. An alternative strategy involves using Treasury Note yields. Historically, risk-averse investors shift funds from stocks to safer bonds and Treasury notes, creating a negative correlation between stock investments and Treasury yields. High bond yields suggest low demand, indicating investors are leaning towards stock investments.

To validate this strategy, we examined the statistical significance of using Treasury yield data. Multiple models, including the SGD Regressor, were trained using S&P 500 OHLC data and 5-year Treasury Note data. Results indicated the models performed better with just S&P 500 data, questioning the reliability of Treasury yield correlations.

Data was collected using Python scripts to export from MetaTrader 5. Analysis showe...
#MQL5 #MT5 #trading #finance

Read more...

The Boids algorithm, conceived by Craig Reynolds in 1986, is renowned for simulating flocking behaviors observed in nature, primarily among birds, fish, and insect swarms. This algorithm employs three fundamental rules to replicate collective behavior: Separation minimizes collision chances, Alignment adjusts movement direction based on nearby entities, and Cohesion brings individuals closer to each other. This simple rule set can generate complex group dynamics.

Applications of Boids include realistic animation in computer graphics, behavior modeling in robotics, traffic management, and more. Additionally, the algorithm has given rise to other swarm intelligence methods like Particle Swarm Optimization (PSO).

Parameter tuning is crucial for Boids. Key parameters include cohesionWeight and cohesionDist for attraction, separationWeight and separ...
#MQL5 #MT5 #algorithms #optimization

Read more...

The Trend Strength Index (TSI) indicator evaluates the current trend's strength. It operates similarly to a simplified auto-regression indicator by utilizing the correlation function. This function computes the Pearson correlation coefficient between two data sets: the closing prices and the bar indices over a specified period. The coefficient, which ranges from -1 to 1, measures the strength and direction of the linear relationship between the bar index and the closing price.

A value close to 1 indicates a strong uptrend with minimal deviation from the mean, while a value near -1 signifies a strong downtrend with small deviations. Values around 0 suggest a possible flat market trend.
#MQL4 #MT4 #trading #forex

Read more...

Write articles for the Algo Trading Community and earn 200 USD for each publication

Do you have valuable knowledge to share or an interesting idea to discuss? Join our authors, share your expertise in trading and programming, and get paid for your publications.

In July, experienced algorithmic traders covered a wide range of topics: from the versatile capabilities of the MQL5 language and basic multi-currency strategies to advanced AI technologies and neural networks in trading.

Last month alone, 61 new publications were released in the Articles section, a record achievement. Here are some of the top articles:

▪️S&P 500 Trading Strategy in MQL5 For Beginners
▪️Build Self Optimizing Expert Advisors with MQL5 and Python
▪️Eigenvectors and Eigenvalues: Exploratory Data Analysis in MetaTrader 5

At MetaQuotes, we fully support knowledge sharing among our community members. We believe that the continuous exchange of ideas can help increase the professional expertise of our users and drive progress in the entire algorithmic trading industry.

Become one of our authors: contribute to the growth of the largest algorithmic trading community and monetize your expertise.

Write an article...

An advanced guide on integrating Python for MQL5 development is now available. This is essential for enhancing programming capabilities, particularly for financial market analysts and developers.

**Python Overview:**
Python, created by Guido van Rossum in 1991, is a high-level, readable, and simple programming language. It supports various paradigms like object-oriented, procedural, and functional programming, and runs on multiple operating systems such as Windows, MacOS, and Linux. Key features include readability, dynamic typing, and an extensive standard library. Popular libraries for data science, AI, automation, and web development are Pandas, NumPy, SciPy, TensorFlow, Selenium, and Flask.

**Integration Benefits:**
Combining Python with MQL5 offers several advantages:
1. Advanced data manipulation through libraries like Pandas and NumPy
2. Machi...
#MQL5 #MT5 #Python #FinTech

Read more...

MetaEditor includes a compiler that effectively manages errors detected during profiling. This tool helped uncover why the previous version failed to display risk-reward rectangles correctly. The issue was not with the code but with the look-back candle value set too high at 5,000 bars by default, slowing down the indicator chart window.

To address this, we reduced the look-back period from 5,000 bars to 1,000 bars, significantly decreasing the amount of data to be calculated. We created a standalone script to handle conditions checked by Buffer 6 and Buffer 7. The script draws necessary risk-reward rectangles and places lines with price labels for the entry price, Stop Loss, and Take Profit.

Now, the focus shifts to developing an Expert Advisor based on the refined Trend Constraint Indicator. Using MetaEditor, we write our EA, incorporating MQL5 fun...
#MQL5 #MT5 #MetaTrader #Trading

Read more...

The Restoring Pull Indicator, developed by John Ehlers, is a momentum indicator that provides clear buy and sell signals. As noted in Stocks & Commodities V.11:10 (395-400), this indicator is one of Ehlers’ earliest contributions to technical analysis. The principle is straightforward: when the indicator line turns green, it signals a buying opportunity. Conversely, when the line turns red, it’s time to sell. This tool is particularly useful for traders looking to capitalize on shifts in market momentum. Understanding its application can enhance decision-making in trading strategies.
#MQL5 #MT5 #Trading #Indicators

Read more...