Algorithmic Trading in Forex: Automated Trading Algorithms

2024.04.10 2024.04.10
Algorithmic Trading in Forex

Oleg Tkachenkohttps://www.litefinance.org/blog/authors/oleg-tkachenko/

Algorithmic trading in Forex means using Expert Advisors (EA) that automatically open and close trades and also calculate the risk level and position volume according to a given algorithm without direct influence by the trader. EAs help increase trading performance, perform almost instantaneous analysis of historical data, and analyze the Forex market using mathematical and statistical models. Trading robots are Indispensable in high-frequency trading strategies, trading on horizontal and vertical volumes, and grid trading with pending orders.

This article explains algorithmic trading for beginners. Let’s get started and study the basics of algorithmic trading in Forex, the principles of automated trading, and the features of Expert Advisors. You will also find practical tips on selecting and employing Expert Advisors in trading in financial markets. 

What is Forex Algorithmic Trading?

Algorithmic trading is a method of trading in financial markets using a special programme or algorithm. Trading robots analyze the state of the cryptocurrency, stock, and Forex market. They search for repeating patterns, place orders, and execute trades without direct human participation.

The goal of algorithmic trading is to automate market analysis and the position management process. In addition, robot trading eliminates the opening of positions under the influence of emotions and helps to optimize the distribution of order volumes across different price levels and so on.

The essence of algorithmic trading is the automation of routine actions.

A trading advisor is software, a code written according to a manual strategy algorithm. In manual trading, you need to search for signals independently and make decisions about entering or exiting a trade. However, a common trading strategy can be translated into code, and then the software will perform all the actions for you.

Trading advisors for Forex algorithmic trading can be divided into two groups:

• Standard advisors. The code contains an algorithm for managing transactions with the ability to calculate position volume and risk level. In accordance with the conditions specified in the code and settings and if all the specified factors coincide, the robot will perform the required actions. If the robot yields a loss, it should be optimized, and the trading algorithm in the code should be corrected.

• Neural networks are algorithms with machine learning based on artificial intelligence. These algorithms find patterns in past price performance and extrapolate them to the current market situation. They analyze the Forex market using mathematical and statistical models and choose the best option—buy or sell. Trading robots are capable of self-learning (machine learning). Such algorithmic trading systems work with thousands of instruments, selecting their best combination.

The second type of expert advisor is often used by institutional investors in scalping, where buy and sell orders are completed in a fraction of a second. This type of Forex trading is also called High-Frequency Trading. Standard advisors can be used in any situation, depending on the algorithm embedded in the code.

Key Takeaways

What is the Difference between Automated Trading and Algorithmic Trading?

Algorithmic Forex trading is a method of executing a large order by splitting it into many small parts. These small orders are placed in the market at a certain period of time and at a certain price using special trading algorithms. The aim of algorithmic trading is to reduce the cost of executing a large order, reduce its impact on the price, and lower the risk of the order not being filled due to the lack of counter offers.

Also, this concept often refers to automatic trading using certain algorithms, which is called Forex automated trading on Forex. 

Automated Forex trading is a process where trading decisions are made and executed using special software or an algorithm that follows specific pre-defined rules or strategies. The goal of an automated trading system is to make a profit in the Forex market using various technical analysis indicators, price action patterns, statistical models, artificial intelligence, and other analysis methods.

This interpretation is considered from the point of view of the essence of the process. Automated trading implies that robots enter and exit trades for the trader. Algorithmic trading means that an algorithm is used to execute large orders with minimal losses by splitting them.

In most sources, the definitions of “automated” and “algorithmic” trading are synonyms that are used as identical concepts. The essence of the modern term Algo trading is making transactions by trading robots.

Algorithmic Trading Strategies

An Expert Advisor is a manual strategy turned into code. Therefore, a Forex algorithmic trading strategy is the same trading system as used in manual trading. Some strategies may seem complicated to novice traders, so they are turned into automated expert advisors. Some strategies that are considered difficult to manually apply and manage are discussed below. 

Trend-following Strategies

A trend-following strategy is a trading system built on the tendency of high and low prices to fluctuate in a certain direction over a long period of time. The goal is to determine the beginning of a trend at the moment of a price reversal or price exit from a flat and enter a trade in its direction. This system is based on the overall market sentiment.

The simplest algorithm of a trend-trading robot for currency pairs:

• It determines overbought/oversold zones based on oscillators. These are the zones in which a reversal with subsequent movement to the opposite zone is most likely.

• It analyzes signals sent by trend indicators. For example, the position of moving averages relative to the price of a foreign currency.

• It identifies potential zones of order congestion that may impede trend movement.

• It calculates relative volatility according to the corresponding indicator. An increase in volatility is the probability of a new trend.

If all signals coincide simultaneously, the EA enters a trade or places a pending order in the direction of the trend.

After an upward movement, the exchange rate of the particular currency pair goes down, breaking through the EMA. The RSI is moving down from the overbought zone. There is a coincidence of two signals that the robot perceives as a signal to open a transaction.

Two points of algorithmic trading:

• This example of currency trading is the simplest. Robots work according to more complex models in the foreign currency market.

• In the settings of some robots you can specify the shift of signals by candlesticks. For example, in this case, two signals appeared on two consecutive bars.

Trend trading is one of the favorite Forex algorithmic trading strategies among traders, institutional investors and hedge funds, differing only in horizon and time frames. Retail Forex traders often look for short- and medium-term market trends – an intraday trend movement, a trend lasting several days. Institutional investors or hedge funds work with trends that last from several months to several years.

Arbitrage Opportunities

Arbitrage is a trading strategy suggesting you make money on the difference in the price of one currency pair in different markets or types of the trading platform. For example, you buy BTC on one cryptocurrency exchange and simultaneously sell it on another, provided the difference yields you a profit.

An arbitrage Forex trader buys an asset where it is cheaper and at the same time sells it where it is more expensive, making money on the price differences over a short period of time. The arbitrage strategy can be spatial, using the difference in the price of one instrument on different exchanges, or it can be time arbitrage. Time arbitrage refers to a profit made on the price difference over time. 

Arbitrage is possible due to market inefficiencies, when the price of an instrument does not reflect its true value, or when there are time delays in the transfer of information between trading exchanges. New quotes have already arrived on one trading platform but not yet on the other. And the difference covers the spread. Arbitrage provides profitable trading opportunities.

Requirements for an arbitrage trader:

• Constant monitoring of the trading platform, exchanges, brokers – monitoring of their tariffs.

• Tight spread amid high market liquidity. The price difference due to technical failures of the trading platform or delays in quotes is small. Therefore, the asset must have a really tight spread; otherwise, the trading strategy will be unprofitable.

• Instant execution of transactions. As soon as a pair of trading platforms is found, you need to immediately make a trade in both directions until the price levels out.

A trader cannot carry out such operations manually, even with dozens of Telegram channels and other providers of recommendations on assets suitable for arbitrage. Arbitrage is one of the few algorithmic execution strategies that only robots can implement.

Index Fund Rebalancing

This strategy is suitable for the long-term stock investor. The idea is to constantly review the portfolio structure and adjust it. Unprofitable shares are sold when the price falls, and profitable ones are bought back.

Peculiarities of rebalancing in the stock market:

• You need to wait out the drawdown.

• You need to correctly calculate the share of an instrument in a portfolio, taking into account the level of risk and following the risk management.

• You need to determine overbought and oversold states of assets.

Manual rebalancing in stock or currency trading is inconvenient for several reasons. Firstly, you should predetermine the period of time when to rebalance your portfolio.  If you do it once a month, there is a risk of selling promising shares during a local correction and buying additional overvalued securities. Revising your portfolio once a year means delaying the sale of an unprofitable asset. Secondly, you can’t calculate the level of risk if you do not know mathematical and statistical tools.

So, an automated system should be useful here. According to the underlying algorithm, it calculates the risk level using several models: Sharpe, alpha, beta coefficients, and so on. Based on them, it determines the optimal ratio of the asset to the overall portfolio. Based on historical data, it calculates the potential depth of the drawdown, after which the instrument should be sold. It also acquires additional undervalued assets. And all this happens not at some fixed time intervals but continuously and at the best prices.

Mathematical Model-based Strategies

Algorithmic Forex trading strategies are based on mathematical and statistical laws. These Forex strategies use standard deviation, variance, correlation, and so on. For example:

• The regression model method uses statistical regression to analyze the relationship between stock prices and other variables.

• The spectrum analysis model is built on non-standard digital indicators that track price noise at different time intervals.

• The Monte Carlo model allows you to generate many random scenarios of market conditions, as well as estimate the probability and consequences of various outcomes.

• Quantum models mean a combination of arbitrage, quantitative analysis, and high-speed trading.

Building such models manually and doing calculations is pointless. The robot does all this, after which it offers the optimal solution for trades based on calculations.

Trading Range (Mean Reversion)

Trading range is an algo Forex trading strategy that uses the price channel as the main indicator to determine entry and exit points in the Forex market. A price channel is a chart formation that consists of two parallel lines or curves that limit price sways within a certain range.

The idea behind the range trading strategy is that the price moves within a range and eventually tends to return to its mean value – the middle of the channel. The further the price deviates from its average value over a particular period of time, the more likely it is to reverse in the opposite direction.

The simplest algorithmic Forex strategy based on the range trading system (mean reversion strategy) for currency pairs looks like this:

• Using standard deviation indicators, moving averages, channel indicators (Bollinger Bands, Keltner Channel), and fractals, the robot determines the support and resistance dynamic levels. 

• A signal to enter a trade is a price rebound from the channel border or an impulse breakout, followed by a reversal and return of the price to the channel. The EA opens a position when the price goes into the range. Pending orders can be used in case a channel breakout turns into a new trend, and the channel expands.

• The lot is calculated in accordance with the specified settings.

• The trade is exited when the price reaches the middle of the trading range. Or the position can be closed in parts: 50% is closed when the price reaches the middle of the channels and the rest is closed when the price reaches the opposite border of the range. 

Points 1 indicate trade entries, and exits are at points 2. You can see trade exits on the opposite side of the range. 

Expert Advisors often use a trailing stop—a stop loss that follows the price. But it requires a VPS server.

Volume-weighted Average Price (VWAP)

The trading model is based on analyzing horizontal and vertical trading volumes:

• Vertical volumes show the trading volume for a certain time. They mean trading volumes on a specific candlestick.

• Horizontal volumes show trading volumes at a certain price level. They mean the number and volume of transactions at a specific price.

Key analysis tools are volume indicators and market depth. The robot can:

Quotes and volumes in the market depth change dynamically. In short time frames, the change can occur within a few seconds. The trader cannot track the market data changing at such a speed, and here, an Expert Advisor is of great use.

Percentage of Volume (POV)

This is algorithmic trading, which automatically determines the transaction volume, which will not significantly impact the price. Placing a large order without counter orders can greatly change the price and increase market volatility. The robot splits the order and places small orders as counter orders appear. Thus, it gradually satisfies the requests of counterparties until the entire order is executed.

Implementation Shortfall

The Implementation Shortfall trading strategy is a portfolio management method that minimizes the difference between the expected and actual execution prices of trading orders. It can also be used for auto hedging. In this strategy, the robot also manages the overall position’s volume, with reference not to the volume of counter orders but to the spread size.

The wider the spread when there is a lack of liquidity, the less favorable the price at which the trader enters a trade. Conversely, it makes sense to gain maximum position volumes with a narrow spread, counting on its further expansion and subsequent sales. When collecting the full volume of a long position with a narrow spread at one time. Buying in parts on a widening spread is a risk of buying an instrument at a less attractive price.

The goal of such trading algorithms is to achieve a balance between execution speed and market impact, that is, the impact of the transaction on the price. It also aims to optimize the volume of the overall position, depending on the level of the current spread, considering the acceptable level of risk.

Beyond the Usual Trading Algorithms

Counter orders algorithm. The idea of the strategy is trading Forex robots against robots. It comes down to identifying the algorithms of large traders and placing counter orders. The robot of an institutional investor, for example, a hedge fund, places orders to buy an instrument at certain price levels. Let’s assume that it is gaining a position in parts not to influence the Forex market. Your robot finds such orders, finds the security cheaper, buys it, and sells it to the robot of an institutional investor. The difference in prices is your profit.

The strategy is a cross between volume trading and arbitrage. Since such transactions are carried out in a split second, automated tools are indispensable.

High-Frequency Algo Trading

High-frequency algorithmic trading is HFT trading, which involves an automatic system opening and closing trading positions within a fraction of a second. The essence of algorithmic trading comes down to making money on the smallest price movement, allowing retail traders to make money on numerous transactions without holding them. This requires compliance with several conditions:

• Spread from 0 points. Speculative trading works if there is almost no commission. Therefore, such robots are launched exclusively on ECN accounts.

• Order execution speed. The average speed in the market is 200-300 milliseconds. The ideal speed for stock trading and trading Forex is considered to be a maximum of 30-50 milliseconds.

The HFT strategy requires a trader to use large computing power. Therefore, high frequency trading is employed mainly by institutional investors with computer access to powerful servers. The strategy’s disadvantage is the costs of regulators and the trading platform.

Front Running

The Front Running strategy implies that the robot places an order to buy or sell an asset before a large order from the market maker, in the expectation or with the goal that the large order will play the role of support/resistance.

First, orders in the market depth are automatically analyzed (instant liquidity). An order is executed if it appears next to the Bid/Ask price and significantly exceeds the average volume of orders in the market depth or the average volume of transactions for a certain time. The strategy is designed so that before large orders are satisfied, the price will rebound several times in the opposite direction.

Front running is also used by scalpers, who make many short-term trades in an attempt to capture small price movements. Algorithms use market depth, so you need a broker who will give you a depth of market of at least 20*20.

Technical Requirements for Algorithmic Trading

1. Computer requirements:

• Processor from Intel CORE i5, i7; AMD Ryzen 5, 7.

• RAM—at least 8 GB. The basic version of almost all modern desktops and laptops is equipped with enough memory to run several advisors.

• SSD—at least 50 GB. This is enough to install a trading platform, advisors, and other algorithmic trading software that requires saving the history of quotes, libraries, etc. An SSD drive is faster than an HDD.

2. Trading software. The operating system must be Windows 10 or higher. MT4/MT5 class platforms cannot be installed on Windows 7.

3. A VPS server is needed to ensure the uninterrupted operation of Expert Advisors when the computer is turned off or there is no connection.

You also need a stable Internet connection (optics, Starlink) with a speed of at least 100 Mbit/s. You should work with a reliable broker who will supply the platform with quotes and data in the market depth without delay.

If we are talking about professional automated Forex trading, you will need equipment similar to server equipment: a processor of at least Intel Xeon Gold 5118 class and Windows Server 2012/2016/2019/2022 (x64).

To trade using algorithms, you need to have knowledge in the field of stock trading and technical analysis. Programming knowledge will also be useful, although this is no longer quite a technical component. It is advisable to have an idea of how code is written or to have someone who can help customize it. You also need an advisor tester and the ability to use it. For example, for EAs in the MQL programming language, MT4/MT5 has a built-in tester. You can also use a separate programme: Fx Blue and Forex Simulator.

How To Choose An Algo Trading Strategy

Now, look at the recommendations for choosing a Forex trading strategy:

1. Consider code and platform compliance. Code written in C# cannot be run in MT4 and MT5. Conversely, an MQL-based Expert Advisor will not be suitable for the cTrader platform.

2. The higher the desired return, the greater the risk to lose money you will have to take. For example, risks increase if you launch several EAs at once or one EA on several instruments. 

3. You must understand what indicators, signals, time intervals, and finance instruments the EA works on. How it manages positions, stop losses, take profits, and other parameters. You should know What each parameter in the settings means and is responsible for.

4. It is necessary to check how the EA has performed in the past and present under different market conditions. You should understand how it reacts to increased volatility, sharp price movements, news, and other factors.

Choosing an EA comes down to the following algorithm: run it in a tester, run it on different assets and timeframes, and try different combinations of settings. Look at what interval, under what conditions, and which EA has the best result.

Additional tips for trading currencies and other assets:

• Don’t rush to take paid Expert Advisors. These may be slightly modified free ones. Most robots are unprofitable not because they are “bad” but because algorithmic Forex traders do not know how to work with them – set them up and adjust them to specific assets and time frames.

• Still, paid EAs have benefits. They should include a history of transactions on a real account. The seller can also help adapt the expert advisor to a specific task and optimize it.

The best option is an Expert Advisor developed using a successful manual strategy.

What are the Benefits of Algorithmic Trading?

In the context of algorithmic trading in the stock market and foreign exchange – slicing a large order into smaller ones – the main advantage is the gradual absorption of counter orders.

Example: You want to buy 1000 shares for 100 USD each. But the sellers only have 400 shares. Seeing the increased demand, the owners of the securities will immediately raise the price, and you will have to buy the securities conditionally at 110-115-120 USD. The algorithm splits your order into several smaller ones. First, the available volume (400 shares) is purchased. This does not cause a change in price since the sellers agreed to the price. The robot then waits for new sellers to appear at the same price.

Thus, you gradually accumulate the volume of 1000 shares at a price of 100 USD. The robot’s role is to instantly react to emerging orders from sellers.

In the context of automated trading systems, the advantages of using Expert Advisors are as follows:

• Reaction speed. A human trader is completely inferior to a robot in terms of transaction speed. Therefore, almost all scalping and high-frequency trading are performed by robots.

• Automation of actions. Expert Advisors can be launched simultaneously on several assets. Robots are capable of real-time monitoring ten trades on ten charts simultaneously, which a trader is unlikely to be able to do.

• This reduces the trader’s workload, which is primarily visual and mental. Instead of monitoring dozens of markets and charts, the trader monitors the news and the deposit. He/she spends no time on anything other than fundamental analysis.

• Diversification of risks. Different types of EAs can be launched on different charts, working at different times, to eliminate the simultaneous load on the deposit.

• Eliminating the influence of emotions. A person under emotion tends to make mistakes—moving stop-loss orders in the hope of a reversal, guided by FOMO instead of logic. The robot is impartial. 

However, these advantages can easily turn into disadvantages.

What are the Risks of Using Algorithmic Trading in Forex?

The risks of using an algorithmic trading system include

1. Influence of fundamental factors. An EA opens a position regardless of what happens in the market. For example, a robot sees a series of signals confirming that an asset is overbought, so it opens a short position. However, following an inflation report, retail traders begin to buy assets en masse under the influence of a fundamental factor. In this case, the EA entered a losing trade.

2. Influence of large market participants (financial institutions, market makers). With large volumes over a short-term period of time, the market maker can deliberately shift the price in the desired direction to collect liquidity and enter the Forex market at the best price. The robot algorithm cannot provide for such actions. This also includes spoofing – robots place orders in a split second and delete them to create the appearance of trading and volumes in the depth of the market. This misleads other robots.

3. Sensitivity to high-volatility markets. These are sudden price fluctuations, which are especially acute for cryptocurrency and foreign exchange markets. For example, a trader, due to their risk appetite, can manually increase the size of stop-loss orders or decide to close a position earlier. However, a robot acts according to a strict algorithm.

4. Lack of flexibility. In manual mode, a trader can change the volume of transactions, the length of take profit, and stop-loss orders to manage risks depending on the market situation. The Expert Advisor calculates the position volume using a given algorithm. Where it is necessary to reduce the volume, it will continue to increase it, thereby increasing the risk. This is especially acute for robots using the Martingale strategy.

5. Code errors. The EA’s coding is not always perfect. Even if conditions are met, it may fail: it will enter a trade at a loss with an incorrect volume calculation, close profitable trades, and so on.

6. Technical errors, power or internet outages—a solution is to keep a mobile application and a VPS server at hand.

Most of these risks can be eliminated or minimized if you follow successful traders and combine algorithmic systems with discretionary trading:

I recommend you read this article to learn the most common mistakes traders make using automated trading.

Forex Algorithmic Trading: Pros and Cons

The table below presents the pros and cons of Forex algo trading.

Conclusion

Algo trading: summary

• Algorithmic trading systems involve making trades with robots or Expert Advisors. Although this might seem complicated for novice traders, it is a good opportunity to develop non-standard trading skills when training on a demo account. EAs can be used in stock, commodity, crypto, and Forex markets.

• An Expert Advisor is software based on a manual strategy algorithm. The robot is launched on a demo or real account without the trader’s direct participation. An EA opens and closes transactions, sets pending orders, and calculates position volume.

• A complex trading system can enter algorithmic trades simultaneously across multiple markets and platforms. Arbitrage and HFT trading are strategies that can be implemented exclusively with the help of robots.

• Expert Advisors allow you to implement strategies based on mathematical modeling, quantum and spectral analyses, and statistical algorithms.

• A trading robot allows successful traders to computerize actions that were performed manually, increasing productivity and eliminating emotions from Forex trading.

• The disadvantage of the advisor is that it is just a robot that works according to a built-in algorithm. An EA operates without intuition, risk appetite, and the prospect of destroying the deposit if there are code errors.

• Before launching any Expet Aadvisor on a real account, it is first checked against the price history in a strategy tester with at least 200-300 transactions.

An Expert Advisor is only an assistant. One click of a button is not enough to start making money. Algorithmic trading with the help of an Expert Advisor requires control. You should constantly monitor trading statistics in comparison with the backtest results, monitoring its work in the period of time of news release. If you learn how to work with an algorithmic trading system, you can significantly increase your Forex trading performance.