Abstract: Escalations in computer science and communication technologies have created new opportunities to ameliorate or extend applications. The recent developments have created room for new trading strategies. Reformations have come into the limelight at both the investment decision level and the order execution level. The article puts emphasis on the question “What is Algo Trading?”. It aims to furnish the reader’s understanding of the why(s) of algorithmic trading. It also throws light on the transformation from manual to automation. As a reader, you’ll know the different Algo-trading strategies put to use by today’s investment firms.

Introduction

The evolution of trading is one of the most significant factors in humanity’s journey. But exactly how has the Stock Exchange evolved over the years? For the first time, Belgium boasted a stock exchange in Antwerp in 1531 where brokers and money lenders would meet to deal in business, government and even individual debt issues. There were no promissory notes and bonds or real stocks. In India, it was 1860 when an informal group of stockbrokers organized themselves as the “The Native Share and Stockbrokers Association”. It formally came into existence as the Bombay Stock Exchange (BSE) in 1875. BSE is presently the oldest stock exchange in Asia. 

The National Stock Exchange (NSE) took form on 4th November 1994 with the exponential growth of trading in India and the world. People traded manually by trading electronically using telephones and computers in past decades. Without automated trading, traders used to collect and analyze market data and make trading decisions based on it. Therefore, there were only humans who could decide to buy or sell stocks based on market data in the past. Over the course of time, a need for, faster, (disencumbered from human emotions) and more accurate methods came into existence. Algorithmic trading turns out to be the appropriate course of action.

Understanding the Concept of Algorithmic Trading

The very first question that arises is what is Algo trading or Algorithmic trading? If we actually do a web search, we see that different regulators’ definitions have slight differences. Different organisation have put their views forward for the question “What is algo trading”. As per SEBI (the regulator for financial markets in India), “any order that comes into play through automation of execution logic is Algorithmic Trading.” MIFID, FINRA, FCA, etc., have slightly different definitions.

In the algorithmic trading definition, regulators emphasize an automated and computerized decision-making process. The process defines individual trading parameters for an order in terms of timing, price and quantity setting, and post-receipt order management without human intervention. For lucidity, algorithmic trading does not actually define transaction parameters. It does not include systems that only handle automatic order routing or order execution confirmation.

According to the “Global Algorithmic Trading Market 2018-2022” report by Research and Markets, if data is to be reliable, the global algorithmic trading market size is projected to grow from $11.1 billion in 2019 to $18.8 billion by 2024, expanding at a CAGR of 11.1 per cent. 84% of trades that happened in NYSE, 60% in LSE and 40% in NSE were done using algorithmic trading. From the looks of it, it seems that every trade will be incorporating algorithms sooner or later. But why is that?

The What & Why of Algorithmic Trading

In trading, it is more difficult to execute large orders due to strong market influence and high signalling risk. One way to overcome this problem is to split the bulk order and distribute execution over time. This minimizes the associated implicit transaction costs or trade using algorithms. When we discuss ‘Algorithms’, the most important mention is of Jim Simons – The man who solved the market.

“A program using automation to implement an order execution strategy defines what is Algo Trading. (also called automated trading, black-box trading, or algo-trading)”. Technically, it uses several mathematical algorithms to determine transactions whose calculation is on current market data and submit & execute orders in the financial markets. In this way, the computer systematically makes the decision to execute each transaction, freeing the transaction from all human emotional influences (fear, greed, etc.) & helping us overcome behavioural financial decisions.

Frequencies of Trading (HFT, MFT, LFT)

Travelling into the past, it was not until the late 1980s and 1990s that Algorithmic trading (completely electronic execution of exchange) started in monetary business sectors and the financial markets. By 1998, US Securities and Exchange Commission (SEC) approved electronic trades, making them ready for mechanized High-Frequency Trading (HFT). Also, since HFT had the option to execute exchanges multiple times (approx. 1000) quicker than a human, it became widespread.

High-Frequency Trading (HFT) is a sort of Computerized Exchanging, the clarification of which we will see ahead. Albeit Algorithmic Trading is one concept of executing the trade, there are various degrees of frequencies (speed) at which it operates in the securities exchange.

Now, there is a particular level of speed at which trading takes place. Profits generated every second are decided by this speed. Here is the explanation of three types of trading, based on their frequency or speed.

High-Frequency Trading (HFT)

High-Frequency trading basically leads to high-speed trades. Simply, the execution of large numbers of orders takes place in a fraction of a second. Consequently, the trading of securities is possible every millisecond in the fluctuating market, making it exuberantly profitable. The HFT strategy was first made successful by Renaissance Technologies LLC. It is an American hedge fund and is among the most successful hedge funds in the world. Renaissance spread its roots in 1982 by ‘the man who solved the market: James Simon‘.

Latency of HFT

 As latency is accountable for the speed of trading, HFT is a low-latency trading practice which implies that the exchanging or trading happens a lot quicker than the opposite reaction in response to market events for expanding profitability.

In the capital market, low-latency is the utilization of algorithmic trading to respond to market events quicker than the opposition to expand the benefit of exchanges. For instance, whenever executing arbitrage strategies, the chance to “arb” the market may just introduce itself for a couple of milliseconds before parity is achievable. What is thought of as “low” is subsequently relative yet, in addition, an unavoidable prophecy.

Numerous associations and organizations are utilizing the words “ultra-low latency” to portray latencies of under one millisecond. However, it is a developing definition, with the measure of time considered “low”, always contracting. Numerous specialized elements sway the time it takes an exchanging framework to identify an opportunity and effectively take advantage of that chance. Firms occupied with low latency trading are willing to contribute impressive exertion and assets to speed up their trading innovation as the additions can be significant.

Interesting Facts About High-Frequency Trading

1. HFT firms are market makers and provide liquidity to the market, which has lower volatility and helps narrow bid-offers spread to make trading and to invest cheaper for other market participants.
2. HFT scaled up its fame in consequence of exchanges offering gigantic incentives for the organizations & firms so as to add liquidity on the lookout.
3. In US & other developed countries, HFT accounts for 70%+ of the equities, and in India, HFT accounts for 33%+ of its financial sector and is growing rapidly.
4. High-frequency funds started to become especially popular in 2007 and 2008 and massive scaling from 2014. Here is a chart of analysis with Google Trends, which indicates that news volumes surrounding HFT spiked in April 2014 to levels substantially higher than any previously.

Medium-Frequency Trading (MFT)

Medium-Frequency Trading systems incorporate all exchanging exercises that don’t need market microstructure examination on one side and altogether rely upon market sway on the opposite side. The main contrast from High-Frequency Trading is the capacity to analyze enormous measures of data utilizing complex calculations and algorithms. Basically, MFT takes a few minutes to a day to place the trade. As a result, it is more moderate than in comparison to HFT. It has higher latency than HFT.

Low–Frequency Trading (LFT)

Low-Frequency Trading is the slowest type of trading and usually takes place in a day to a couple of weeks. The latency time in LFT is much higher than HFT & MFT.

Strategies of Algorithmic Trading

We can now derive that algorithmic trading requires strategies for making the most beneficial decision, Algo trading strategies are several sorts of ideas for directing the most profitable algorithmic trade. The idea for the formulation of the pathway is in a unique manner. It starts with retrieving real market information feed from the trade. With the pre-characterized lump of rules or rationale, it generates a trading request. Algo trading strategies play a crucial role is making a profitable decision.

There is a pre-defined manner in which each algo trading strategies work to furnish the trader with unerring execution of algorithms. Here is the explanation for the most popular algo trading strategies :

Market Making Strategies

This algo trading strategy is generally used by market makers, who are generally large institutions or firms. Market makers facilitate trade orders on a macro scale for buying & selling. The reason behind them being large is that an immense amount of securities are involved in the same. This strategy helps in injecting liquidity into the markets.

In this process, market makers buy at the best bid in the current market situation and sell at the best quotation for a specific number of securities. When a purchaser gets an order, the market maker sells the offers from its own inventory and finishes the request. Subsequently, it guarantees liquidity in the monetary business sectors, simplifying it for financial backers just as dealers trade. This summarizes that market makers are extremely critical for sufficing trade. One important note is that though the market makers buy & sell according to the current market scenario, they refrain from making trade in case of extreme volatility.

Momentum Strategies

When we base our algo trading strategy on the existing market trend and make utilization of statistics to determine whether the existing trend will continue or not to increase profitability accordingly, it is called Momentum Strategy. In this algo trading strategy, traders seek to buy high and sell higher to make the stocks profitable. Now one question arises- Why does this Momentum work?

It works because traders and investors exhibit a large number of emotional decisions when deviation from the mean is high. As a result, some intelligent investors make gains due to behavioural biases and the emotional mistakes of others.

However, it isn’t peanuts as it carries more volatility than in comparison to most other strategies and endeavours to capitalize on market volatility. It requires pertinent diversification and thoroughgoing risk management techniques to avoid losses in times of grave crashes.

Arbitrage Strategies

This strategy is a venture methodology where an investor all the while trades a financial instrument or an asset in different markets to take advantage of a value contrast or mispricing and produce a benefit. While price differences are ordinarily little and fleeting, the profits can be great when increased by a huge volume. This strategy involves no risk as you execute multiple trades simultaneously on one asset to book profit. This strategy is event-driven as pricing inefficacies happen during corporate events like bankruptcy, acquisition, merger, etc. Arbitrage strategy is commonly leveraged by hedge funds and proprietary traders.

Now, as this misquoting in prices exists only for a short duration of time as prices in the market are adjusted quickly, such tracks are made easier using automated machines and statistical operations. Therefore statistical arbitrage is heavily reliant on computer models and analysis and is considered one of the most rigorous approaches to investing.

Machine Learning Strategies

In Machine Learning based trading strategies, algorithms and patterns are studies that computers follow to trade in accordance with the market data. This strategy utilizes algorithms to foresee the range for extremely transient value developments at a specific certainty interval. The advantage of utilizing Artificial Intelligence (AI) is that people foster the underlying programming, and the artificial intelligence itself fosters the model and further develops it after some time.

There are countless assets that depend on computer models worked by information researchers and quants, yet they’re normally static. For example, they don’t change with the market. Machine learning trading models are highly time-efficient as they can break down a lot of data at high speed and indulge in betterment themselves through analysis. There are AI models available which have high-grade techniques, including Evolutionary Computation & Deep Learning which can run across thousands of machines.

Why Should Retail Investors/HNIs Do Algorithmic Trading?

Well, the fact that retail investors & HNIs (high net-worth individuals) are the ones who had stayed denied algorithmic trading for quite a while. Be that as it may, presently, retail investors are showing interest in algorithmic trading since some revolutionary organizations & brokers like TD Ameritrade are supporting retail investors/HNIs.

Interestingly, retail investors should comprehend that to get into the world of algo trading. They need to have sound information on speculation, investment and algorithmic trading. Albeit, not taking part in algorithmic trading may prompt an effect on the retail investors on the grounds that algorithmic traders might have an advantage over manual traders in the market. In the financial markets, algo trading escorts sundry benefits.

Benefits of Algo Trading to Retail Investors and HNIs

Increased execution speed

The fundamental explanation is assuming you are trading a technique that is beneficial for you. You should be provident enough to speed up execution for getting the profitable trades going rapidly. In trading, you come out profitable just when your successes make up for your misfortunes. That as well, enough in order to represent your endeavours and expenses. Algorithmic trading is a method for doing likewise.

Trading decisions

Traditionally, we have seen retail investors trade according to their ‘gut feeling’ about the market’s future, thereby making decisions affected by emotions. This ‘gut feeling’ isn’t rational and compels them to take behavioural financial decisions and often places the traders under heavy losses. Algorithmic trading follows pre-decided entry & exit rules which prevent such emotional trading decisions and hence avoidable losses.

Market reach 

One of the principal reasons algorithmic trading has been acquiring prominence is that it permits traders/investors to assemble techniques quantitatively. Moreover, it utilizes demonstrating strategies to have the option to oversee risks. This further empowers them to trade instruments, for example, options and derivatives, which are generally excessively unpredictable for retail players/HNIs.

Elimination of constant market monitoring

In algo trading, monitoring of the market, decision making & execution of trades can be done by algorithms. As a beneficial result, there is no need to monitor the market during trading hours continuously.

Real-time quantitative analysis

Using algorithms, we can do real-time quantitative analysis by utilizing past data to help traders in dissecting strategy’s exhibition concerning profit and loss, just as some well-known performance statistics like Sharpe ratio, alpha, beta, and so on. The capacity to backtest and evaluate the strategy’s return over hazard assists the traders with gaining from their own fallacies in a recreated climate prior to running the strategy in live financial markets.

Future of Algorithmic Trading

India gives a decent opportunity for algorithmic trading due to the amendments seen over the last few years. It is because of the various factors like colocation offices and modernized sophisticated technology at both the significant exchanges; a savvy order steering framework; and well-established & liquid stock exchanges. The figure below shows the possible future of the global algo trading market:

The Top Algo Trading Platforms In India

India is witnessing a rise in the demand for algo trading platforms that are convenient to comprehend and reliable for long-term use. The section below lists different algo trading platforms with distinctive characteristics.

Presto ATS

It is a dynamic & versatile automated algo trading platform that Symphony Fintech develops for trading in almost all asset classes.

The three modes it offers are (i) Live training, (ii) Paper training & (iii) Backtesting.

Presto has a host of APIs to be associated with the Indian exchanges. The choices are shifted for APIs with accessibility in Java, C#NET, and in HTML. Despite the fact that it is utilized by institutional investors, it is acquiring prominence among retail traders/HNIs as well. The expense of authorizing the product for a yearly membership is Rs. 25,000 with the choice of single and multi-exchanging accounts.

ODIN

It is a multi-exchange, multi-segment front-office algo trading platform. ODIN with its dual features also plays the role of a risk management system. This algo trading platform was developed by 63 Moons, and it offers Orders Management System (OMS), risk management & third party API integration.

Omnesys NEST

It is a top-tier algorithmic exchange platform that is fit for executing a few techniques like basket trading, order slicing, 2L and 3L spreading. This algo trading platform is very adaptable in its activity and engages intermediaries to exchange across different resource classes like Equities, Derivatives, Currency and Commodities. Despite the fact that it is very amazing as far as its devices, it is, for the most part taking into account institutional dealers and, subsequently, isn’t utilized by retail brokers.

AlgoNomics 

It’s an algo trading platform that caters to institutional investors or investment banks and individual investors. The algo trading platform was developed by NSEIT, which is a subsidiary of NSE (National Stock Exchange). It claims to offer ultra-low latency to its users, ergonomics and furnishes support to all market classes inclusive of Equity, Derivatives, and currency derivatives.

The undeniable benefit is that a retail investor can make their algo trading strategies in another climate however, utilize the dealers API to put in live requests in the market. Simultaneously, one ought to consider the expense related to utilizing the API just as the overall downtime, assuming any when you utilize the API.

The market demands for a contemporary time relevant algo trading platform that can address the intricacy of algos dependent on Artificial Intelligence (AI) and Machine Learning (ML). The reception of ML empowers frameworks to help execution processes. It recommends which Algos to utilize and the specific parameters most appropriate for a given goal. Algos will keep on assuming a significant part of the eventual fate of trading as market participants endeavour to track down better approaches to automate their work processes.

References

1.)https://blog.quantinsti.com/algorithmic-trading-india/
2.)https://www.quodfinancial.com/future-of-algorithmic-trading/
3.)https://faculty.fuqua.duke.edu/~charvey/Teaching/BA453_2005/II_On_Jim_.pdf
4.)https://math.berkeley.edu/~berlek/pubs/bloomberg.pdf
https://www.researchgate.net/publication/345319146_Algorithmic_Trading_and_Strategies
5.)https://www.econstor.eu/bitstream/10419/144690/1/860290824.pdf
6.)https://blog.quantinsti.com/algorithmic-trading/
7.)https://blog.quantinsti.com/algorithmic-trading-retail-traders/
8.)https://www.investopedia.com/articles/active-trading/101014/basics-algorithmic-trading-concepts-and-examples.asp
9.)https://academic.oup.com/rfs/article-abstract/31/6/2184/4708266
10.)https://www.prnewswire.com/news-releases/algorithmic-trading-market-to-record-6-cagr-by-2025-witnessing-emergence-of-algotrader-ag-and-argo-se-as-key-market-contributorstechnavio-301359283.html