MQL5 Algo Trading

In the realm of MQL5 development, efficient code management is critical, especially when dealing with repetitive tasks across multiple projects. A powerful strategy to streamline this process is by utilizing code libraries, particularly .ex5 libraries and DLLs, which are essential for developing Expert Advisors, custom indicators, and scripts on the MetaTrader 5 platform. Creating your own MQL5 libraries not only accelerates the development cycle by reusing existing code but also protects your source code through encapsulation, ensuring that functionalities can be shared without exposing the underlying code implementations. Function libraries (.ex5 files) allow for modular design, enhancing code reusability and maintenance. For more advanced integration, MQL5 developers can leverage third-party C++ libraries or expand functionalities with .NET Libraries and Python modules, seamlessl... Read more...

In the evolution of multi-currency EA development, a transition into a more sophisticated operational mode has been noted. Initially, the CAdvisor class facilitated the aggregation of multiple trading strategies into a single array, which enabled simultaneous operations but introduced complexities, such as the need for smaller position sizes and the possibility of canceling out opposing positions. To address these issues, a significant restructuring was implemented. In the updated model, direct market position openings by individual strategies were removed. Instead, strategies now manage virtual positions while a dedicated receiver (CReceiver class and descendants) handles the actual market transactions. This change ensures that even strategies contributing minimal virtual volumes can influence the market, enhancing operational inclusivity and potentially reducing swap costs and draw... Read more...

In developing efficient trading strategies, it's essential to incorporate comprehensive market analysis and predictive techniques for financial instrument movements. These movements typically exhibit correlations with other financial assets and macroeconomic indicators, revealing patterns akin to regulated traffic dynamics, where various elements interact within defined rules, albeit with inherent unpredictability due to individual actions. The ADAPT method, initially introduced for navigating autonomous vehicles, has been effectively adapted for financial markets. This technique involves analyzing past data to predict future trajectories of multiple agents by dynamically learning weight adjustments. It employs a vectorized scene representation to model interactions among agents and map elements which aids in accurately forecasting movements within the financial landscape. Key eleme... Read more...

Expert Advisor developers often encounter the need for their robots to manage multiple orders simultaneously. Achieving this requires the ability to track the total number of open orders and distinguish among them by type. A practical solution involves using the Comment() function, allowing for real-time visual tracking of the order count managed by the robot. This approach is illustrated in a clear and concise code, complete with descriptive notes to aid in implementation. This kind of capability is crucial for developers looking to enhance the functionality and efficiency of their trading strategies. Read more...

Understanding the application of Hidden Markov Models (HMMs) in financial series heralds a significant leap in predictive modeling. HMMs classify temporal data points into hidden states based on the transition probabilities and emission probabilities, making them ideal for financial time series marked by noise and non-linearities. This primer details the use of Python's hmmlearn module to construct an HMM, alongside an integration example with MetaTrader 5 via both Python and MQL5. Key aspects of modeling with HMMs include defining the number of latent states, setting initial state probabilities, transitioning probabilities matrix, and defining emission probabilities. Techniques such as the forward and backward algorithms are used for likelihood estimations across observed data sequences, while the Viterbi algorithm assists in determining the most likely hidden state sequences. For... Read more...

A trading strategy focusing on the EURUSD pair during the M15 timeframe has been outlined for traders interested in technical analysis. The strategy utilizes a set of three indicators: EMA with a period of 34, CCI with an MA period of 50, and the MACD set to default MT5 settings. For executing BUY orders under this strategy, the conditions are specified as follows: the closing of the candle above the EMA, the CCI indicator crossing the 0 level moving into the positive zone, and the MACD indicator positioned below 0 while completing a bullish crossover. Conversely, SELL orders should be considered when the candle closes above the EMA, the CCI indicator surpasses the 0 level moving into the positive, and the MACD indicator remains below 0, indicating a bearish crossover. This systematic approach provides a clear framework for traders looking to harness the combined predictive powers ... Read more...

In the realm of intraday trading with multiple instruments, the simultaneous balancing of risks associated with each symbol's tick price and average daily volatility is paramount. Addressing this, one approaches the establishment of a proper balancing mechanism that serves to normalize profits and safeguard the trading portfolio. The tick price, defined by the minimum price fluctuation within a currency, varies notably between instruments, demanding precise considerations during trade setups. For instance, EURGBP and AUDNZD exhibit different tick values reflecting volatility and trading dynamics. Moving beyond simple price movements, average daily volatility adds another layer of risk assessment. Trading decisions require a solid understanding of how intraday price fluctuations can affect trade outcomes. Differences in volatility like those between EURGBP and CHFJPY significantly in... Read more...

Machine learning in electronic trading has evolved to enhance strategy performance, particularly through focusing on technical indicator values rather than direct price forecasts. A study underscores the limitations of achieving over 50% accuracy in security price prediction. By switching the focus to technical indicators, accuracy can improve up to approximately 70%. This approach utilizes comprehensive data fetched and analyzed using Python and MetaTrader 5, demonstrating a more reliable method by fully accounting for variables influencing technical indicators. Analysis reveals a stark contrast in the effectiveness of predicting price movements versus technical indicators. Modelling focuses on factors fully observable within technical indicators, fundamentally boosting prediction accuracy. This methodology leverages historical data processing and Principal Components Analysis (PCA)... Read more...

The indicator presented provides a precise implementation of John Welles Wilder Jr.'s Original Average True Range (ATR) from his seminal 1978 publication, New Concepts in Technical Trading Systems. It utilizes Wilder's moving average, specifically the smoothed moving average (SMMA), for greater accuracy compared to the commonly used simple moving average (SMA). Notably, this indicator is configured with a default period of 7 rather than the traditional 14, aligning closely with Wilder's specifications. The coding is versatile, allowing compilation in both MQL4 and MQL5 environments, ensuring broad compatibility. For developers interested in accessing the source code, it is now available under the "Public Projects" tab of MetaEditor, labeled as “FMIC”. This resource offers a valuable tool for those looking to integrate a foundational technical analysis element into their trading stra... Read more...

The invasive weed metaheuristic algorithm stands as a notable technique in the field of optimization algorithms. Originating from the resilience and adaptability of weed species, this approach simulates weed population dynamics to solve optimization problems effectively. This algorithm is especially relevant for scenarios where solutions need exploration across a broad and varied landscape, akin to weeds spreading across a field. In technical terms, this algorithm initializes with a random distribution of 'seeds'—potential solutions—across the problem space. Each seed grows into a 'weed', where its fitness is calculated based on the optimization criteria. Weeds with higher fitness scores produce more seeds, thus propagating their traits more extensively in the next generation. The process iterates, continually refining solutions until optimal or near-optimal solutions emerge. This m... Read more...

An implementation of the original Average True Range (ATR) indicator, as developed by John Welles Wilder Jr. and detailed in his 1978 book, "New Concepts in Technical Trading Systems," is now available. This version employs the Wilder’s Moving Average (SMMA) rather than the Simple Moving Average (SMA) commonly used in standard ATR indicators. Here, the period set by default is 7 days, aligning with Wilder's original specifications instead of the commonly used 14-day period. The indicator is compatible with both MQL4 and MQL5, thanks to the use of conditional compilation. Additionally, the source code for this and other related tools can be found under the 'Public Projects' section in the MetaEditor, grouped under the project name "FMIC." This allows developers and technical analysts to access and integrate these tools in their trading systems effectively. Read more...

In the realm of automated trading, understanding triggers—though complex—is crucial for developing effective trading strategies using Expert Advisors (EAs). Triggers, which signal various trading actions such as opening or closing positions, are integral in navigating the volatile trading environment but they come with inherent risk. Issues of reliability make it imperative to approach triggers with caution to avoid missed opportunities or substantial losses. Transitioning from fundamental order systems discussed previously, the attention now shifts to advanced implementation techniques for EAs, specifically focusing on manual interaction and event-driven strategies. Among these methods, manipulating mouse and keyboard events to generate trading orders presents its own set of challenges. The simplicity of such methods is deceptive; they necessitate a high level of precision in coding... Read more...

The Relative Strength Index (RSI), a robust tool created by John Welles Wilder Jr. and detailed in his 1978 book "New Concepts in Technical Trading Systems," is available in a new implementation. The updated version of this indicator features a revised equation that enhances computational efficiency while maintaining the original mathematical integrity. Additionally, the code is compatible with both MQL4 and MQL5 thanks to conditional compilation. For interested developers, the source code for this and other tools can be found in the "Public Projects" section of MetaEditor, labeled under "FMIC." This resource is particularly useful for those involved in financial markets and algorithmic trading. Read more...

In advancing the development of expert advisors (EAs) for trading platforms like MetaTrader 5, the integration of manual controls within automated systems remains a topic of significant importance. The interaction between automated strategies and manual intervention can be finely tuned by employing specific programming techniques. This approach ensures that traders can still exert personal control over the trading process when necessary. For programmers transitioning from general programming to the MQL5 environment, understanding how to manipulate order limits and implement them visually on trading charts is crucial. Especially when these orders are placed manually within an EA context, it is essential to display such limits clearly. This is not just about enhancing usability but also about maintaining the integrity of the trading strategy by preventing accidental order placements. ... Read more...

The Relative Strength Index (RSI), a pivotal tool initially formulated by John Welles Wilder Jr. in his 1978 publication, "New Concepts in Technical Trading Systems," has been adeptly adapted for enhanced computational efficiency while retaining identical mathematical outcomes. The adapted code supports conditional compilation, enabling compatibility with both MQL4 and MQL5 environments. For those interested in further examination or usage, the source code for this and other similar implementations can now be found under the "Public Projects" section of MetaEditor, labeled "FMIC." This initiative provides developers with easily accessible tools to integrate and implement in their trading systems. Read more...

Conditional Generative Adversarial Networks (cGANs) offer a refined approach in neural network training where both the generator and discriminator are conditioned on external data. This variant deviates from typical GANs by using labeled data to steer the generative process, enhancing the application range from image synthesis to sophisticated tasks like financial time series forecasting. Unlike unconditioned GANs that generate outputs from random noise, cGANs utilize specific input types that align closely with the training data, thus producing more targeted and relevant outputs. This specificity is invaluable in fields where the type of generated data needs to conform closely to existing patterns, such as in financial markets where precision is crucial. The technical setup for cGAN typically includes a main generator network that creates data based on conditioned inputs, and a dis... Read more...

In the evolving realm of algorithmic trading, the emphasis on efficiency has propelled the utilization of optimization algorithms to a crucial status. These algorithms are key to developing trading strategies that not only meet desired standards of profitability but also minimize risks. Central to the practical application of such optimizations is the concept of self-optimization in Expert Advisors (EA). This process entails the adaptation of trading strategy parameters in response to market changes, optimizing for elements like profit maximization or risk reduction. An effective system for EA self-optimization involves multiple steps starting with extensive data collection and analysis, followed by goal setting and the application of appropriate optimization algorithms. Post-optimization, rigorous backtesting and validation ensure the strategy's ongoing relevancy against current mar... Read more...

This Expert Advisor (EA) operates as a sample multicurrency system for Metatrader 5, utilizing basic candlestick logic on daily charts (D1) to determine entry points. The simplicity of its design is intended to assist other developers in learning and understanding the fundamentals of EA construction and application in financial trading environments. This tool is especially useful for those looking to implement and test multicurrency strategies in a simulated or live trading scenario. Read more...

In modern trading, multiple methods are utilized for building trading systems, categorized into graphical analysis, computing systems, and news systems. Each method has its own specific tools and approaches. Graphical analysis, favored for its simplicity in placing orders based on visible chart patterns, allows traders to create visual reference points based on price and time. Computing systems use a range of automated tools like EAs, harnessing indicators and neural algorithms to inform trading decisions. News systems analyze broad market trends through global news and insider information, influencing trading strategies. Traders commonly select one primary method while employing others as supplementary tools. Combining these methods can provide a comprehensive view of market conditions, aiding in more informed decision-making processes in trading strategies. Efficient trading rel... Read more...

The latest update of MetaTrader 5 build 4380 introduces important improvements and fixes: • Added Alt+X hotkey for experts list and fixed potential Live Update issue in the desktop platform • Fixed Bitmap object bugs in the strategy tester • Introduced new requirements for the MQL5 Cloud Network: agents running in virtual environments and with no AVX support are not accepted anymore • Improved one click trading panel on the chart in the web platform Read more...