Algorithmic Trading Pros & Cons

Algorithmic trading also goes under the names of black box trading, algo trading, and automated trading. However, it is not the same as automated trading because it can be done manually too.

In algo trading, traders use computer programs or robots that follow specific instructions or algorithms to purchase trades. These instructions are based on mathematical models such as time, quantity, and price. Theoretically, such trades are capable of generating profits at a frequency and speed that is not possible for human traders to achieve.

For example, an algo trader can feed the following two instructions into a computer program:

• Purchase 50 shares of a particular stock when its 50-day moving average becomes more than its 200-day moving average.

• Sell the same shares when the 50-day moving average becomes less than the 200-day moving average.

The computer program that has received these instructions now becomes capable of automatically monitoring the price of the stock. It will automatically purchase and sell the shares when the pre-defined criteria are met.

The trader who has created the above-mentioned computer program does not have to manually monitor the stock or manually purchase or sell the shares. His computer program will automatically obey his instructions.

Algorithmic trading is also high frequency and high-speed trading as it enables the placing of multiple orders across multiple markets at high speeds by robots that have been previously programmed on the basis of multiple criteria.

The concept finds an application in several varieties of investment and trading activities, such as the following:

• Insurance companies, mutual funds, pension funds, and mid-to-long term investors use algorithmic trading to buy a large number of shares to avoid influencing stock prices.

• Speculators, brokers, and short-term traders use this system to create adequate market liquidity for sellers.

• Followers of trends, systematic traders, pairs traders, and hedge funds incorporate their trading strategies into computer programs and then let these programs automatically trade on their behalf.

Advantages of Algorithmic Trading

Algo trading has a number of advantages, as follows:

• Trade orders are placed instantly and accurately

• Major changes in price are avoided by timing trades correctly.

• Algorithms can execute trades at favorable prices.

• Transaction costs are significantly reduced.

• Multiple market conditions are checked simultaneously and automatically.

• Risks of manual errors are significantly reduced.

• Traders can test algo trading using existing real-time and historical data to find out whether it is a proper trading strategy.

• Mistakes caused by psychological and emotional issues are reduced.

Drawbacks of Algo Trading

Anything that has advantages should also have disadvantages. The same holds true for algorithmic trading as well.

Here is a set of drawbacks of algo trading:

1. Technical Know How – Algo traders need to be technically sufficient. In addition to a computer with Internet connection, they need to know how to use programming languages.

2. Loss of Control – If you are not careful, you can easily lose control over your robot. Your program may function in ways you never intended it to, leading to financial losses that you cannot control. You have to backtest your robot thoroughly before getting it to work for you.

3. Dependency on Technology – You have to heavily depend on systems and technology that may be out of your control. For example, if you save your code on your computer instead of a server, a network error could prevent your robot from functioning. If this happens, you will lose some great market opportunities.

4. Requires Monitoring – You cannot actually relax and expect your program to work for you. All computer codes, programs, and robots require careful and continuous monitoring.

Algorithmic Trading Strategies

Algo traders commonly use the following strategies:

1. Following Trends – Algo traders who use this strategy only have to follow trends in channel breakouts, moving averages, technical indicators, and price levels. They don’t have to make any predictions. They initiate trades based on how frequently a desired trend occurs.

2. Percentage of Volume (POV) – Traders implement algorithms that continue to send partial orders till the trade order is completely filled. This is done according to the trade volume and participation ratio within the market.

3. Mathematical Models – Traders use a variety of options by using mathematical models that have been tested and proven.

4. Mean Reversion Strategy – Algo traders who use this strategy implement algorithms based on the assumption that asset prices touch highs and lows for a temporary basis before reverting to their average value. Traders create algorithms on a defined price range so that trades can be automatically placed when asset prices break in and out of their definite price range.

5. Implementation Shortfall – This strategy has been designed to minimize execution costs by trading off real-time markets. It increases targeted rate of participation when price movement is favorable and decreases it when price movement is not favorable.

6. VWAP & TWAP – According to the volume weighted average price (VWAP) strategy, traders break up large orders into small chunks before releasing them into the market in order to execute the order when it is closest to the VWAP.

Followers of the time weighted average price (TWAP) strategy release smaller chunks of larger orders at regular time intervals in between a start time and an end time. They want to execute the order when it is closest to the TWAP between the start time and the end time so that the market impact can be reduced.

Guide to Building Your Own Trading Algorithm

Before building an algorithmic trading robot or creating your own trading algorithm, you must understand what exactly it is and how it works.

You must understand that you are creating a piece of code that is capable of automatically creating buy and sell signals in a variety of financial markets. The code should, therefore, include entry rules, which determine when the robot should buy or sell a trade; exit rules, which determine when to close existing positions; and position sizing rules, which define the buy or sell quantities.

This is what you will require:

• Computer

• Reliable internet connectivity

• Mac or Windows operating system

• MetaTrader 4, a trading platform that incorporates the MetaQuotes Language 4 or MQL4, which is used to write the code for trading strategies

• Instead of MT4, you can use any software program to create your algorithm.

• Sound algorithmic trading strategy that will form the core of your algorithm

Now that you have created your algorithm, you must test it. The process of testing your code to make sure that it is capable of functioning as you intend it to is called backtesting. You also have to continuously monitor your code to make sure that it still functions as desired.

You must also remember to incorporate some risk management tools into your algorithm so that you can successful overcome any unpredictable risks.

Algorithmic Trading Examples

• Execution Algos – These are algorithms that are capable of accepting requests from seasoned investors on what they would like to trade and figuring out the best ways to place those trade orders. They are machine-learning strategies that can examine stock market data to determine intra-day price movements. Long-term investors can use them to reduce portfolio management work as they are high frequency intra-day trading strategies.

• Sentiment Based Algorithms – These are algorithms that are based on public sentiments.

• High Frequency Algorithms (HFT) – This example of algorithmic trading is characterized by high turnover, high speed, and high ratios of order to trade. They leverage electronic trading tools with high frequently financial information. To put it very simply, HFTs use the latest technology and computer code to trade securities at breakneck speeds.

There are plenty of other examples and you will find them on sites such as Trading View and Quantiacs.

Does Algo Trading Really Work?

Technical analysis has several applications and algorithmic trading happens to be one of them. So, it works for everybody, even retail traders although they avoid it thinking that it is something complicated that it is out of their reach.

However, the simple truth is that anybody who understands the fundamentals of algorithmic trading can easily build an algorithmic trading system. Algorithmic trading accounts for 70% of the total trading volume in the US. In developing countries, it accounts for 40% of the total volume.

At the international level, algorithmic trading is done more by professional traders and financial institutions than individuals. Although the technology and skills required for algo trading can be learned, it is still complex and discourages a number of retail traders. It is, therefore, mostly used by option traders, professional traders, hedge funds, scalpers, and arbitragers. To become a successful algo trader, one has to understand programming languages, statistics, and the financial markets. So algo trading is something that truly works.