Why We Shouldn’t Ban Algorithmic Trading ?

Why We Shouldn’t Ban Algorithmic Trading

On Friday, 20th April 2012, two mysterious events occurred on the National Stock Exchange (NSE). In the morning, Infosys futures crashed over 20% and quickly recovered back to the original level. In the afternoon, just before 2:30pm, Nifty futures crashed 6.7% from the 5,350 level back down to 5,000, and then nearly instantly recovered back to 5200. Both crashes were blamed on algorithmic trading.

Program trading has been blamed for “flash crashes” for nearly 25 years. In October 1987, US markets took a nose dive on a single day and for years, the blame game went on with the primary suspect being program trades. This is understandable. Since a computer can trade with much faster speed than a human, it can set off a spiralling price change by continuously buying or selling with no real control. Why then, should we even allow algorithmic trading?

Program trading can provide for great trading opportunities with less human error. Much of the arbitrage that used to happen in Indian markets was manual. To bridge any potential difference in prices between the NSE and BSE (the two largest stock exchanges in India), arbitrageurs would use two computers, manually entering a buy order on one and a sell order on another. The speed of the operator was his biggest skill, so the dealing room would resound with a cacophony of keyboards when an opportunity arose. The problem? A human can only look at so many opportunities, so many price differences remain. A slight error on that keyboard (F2 instead of F1) can result in a large loss. The operators cost money in terms of computers, real estate and benefits. You could eliminate much of these by using a computer to do exactly the same thing.

Program trading can curtail broker front-running and impact costs. Often, when a fund would have to take large positions, their brokers would put their own buy orders earlier, so that the large purchase from the fund would give them a great profit. In India, much of the volume is made up of the top 100 stocks. After that, stocks trade less than 20 cr. a day. For a mutual fund to buy about 25 cr. ($5 million) in a lesser known stock, its size will immediately drive up the price and a broker is quite likely to front-run their purchase. Using an algorithm instead will allow the purchaser to spread their purchase over several days and several brokers, hunting for volume slowly over time.

Algorithms can also provide liquidity where there isn’t otherwise any. For you to purchase a stock, there needs to be a seller in place. Many stocks don’t have the kind of interest from either investors or traders. Of the near-1,500 stocks traded on the NSE, more than 1,000 trade less than 50 lakh (Rs. 5 million) a day. The spread between the buy and sell prices on the exchange may be too wide; a typical market maker provides the liquidity that allows you to buy or sell at a reasonable cost. Market making operations used to be manual earlier; they are now run through algorithms.

Finally, large orders (greater than a few crores in value) are usually blocked by stock exchanges, assuming there has been a fat finger trade or a mistaken entry. Yet, large deals must take place when they must — if an investor decides to exit a large holding in a stock, they might use an algorithm to send in orders in allowable chunks.

Algorithmic trading, however, comes with its own set of problems. A rogue program can place orders continuously and take the entire system down. During Mahurat Trading in October 2011, such a program created a ruckus in the BSE, so much that the exchange canceled all trades made on that day to avoid a payment crisis.

Even if an algorithm splits large orders into parts that stock exchanges let through, the input itself may be faulty (an extra zero for instance) which means the algorithm does exactly what the order limit was designed to restrict: the fat finger trade. This, they say, is what happened with the Infosys order on Friday, 20th April 2012.

With such large orders, stop-losses can get triggered, creating another spiral. Some traders place a protective stop (in simple terms: “Sell-If-The-Price-Falls-To-X”) way below market prices; such stops get taken out when such large orders come by and those investors that sell feel disappointed when the market rebounds immediately. But that will happen even with large “manual” orders; algo trading is only a convenient scapegoat.

That steep falls are only engineered by automated trading is also suspect. The market has “circuit” limits which shut down trading when the index moves over 10%. After the 2009 elections, when markets moved up 10% in a short while, algorithmic trading wasn’t blamed; neither was it when markets crashed 10% in October 2009. The feeling was that the moves were “justified” since there was news behind it (The election results and a Lehman bankruptcy impact respectively). Since there was no “reason” on Friday, computers must have been to blame.

This is just witch-hunting. Computers do exactly as they have been programmed to do, and there will be large errors if they aren’t monitored properly (as manual traders must be). The regulators and the exchange must investigate each such case, and indeed, it has turned out it was more the human input that caused the error. Every rogue trader or trading program must be found and punished. Surveillance needs to get much more sophisticated to detect misbehaving automated trades. Algorithms already use a different code when entering trades; a series of checks can be run whenever required to see if any rules were violated. Some of this cost needs to be borne by the algo-trading community, by fees like a per-transaction or per-order fee payable to SEBI.

But we can’t go around demanding bans on algorithmic trading just because of a flash crash. Knee jerk reactions like that will hurt legitimate players or put them at the mercy of their brokers, and that is plain wrong.

Algorithmic Trading- Is it the New Artificial Intelligence for Investors?


Algorithmic Trading has come a long way, from reading the tick data days to being latest buzz on the trade markets. Has your firm captured the essence of this new artificial intelligence?

The black-box trading was a hush-hush scenario until a year ago, as only a few trading firms offered them to their clients.Nevertheless, today, you can receive a bevy of free algo trading. Therefore, when a firm announces the discovery of the next generation of trading platform, the financial services industry ears perk up. Here it is about interpreting newsflow the moment it is delivered from the horse’s mouth.

“A client depends on a sell-side firm to execute its order on its behalf according to certain benchmarks. But, when it comes to a strategy, the Algorithmic Trading system actually generates the orders. Some hedge funds do this.”

For level-one sell side firms, automated trading has become a crucial part of their existence. These firms employ algorithm trading as the basic strategy to gain momentum in the markets across the globe.

How does this artificial intelligence work? Algorithmic trading is standardized by Volume-weighted average price (VWAP) it divides the total value of trades by the total volume over a period. Interestingly, European-banking conglomerate UBS utilizes algo trading for its 40% clientele. Other giant corporations who are testing the unique capabilities of robo trading include JPMorgan Chase, Credit Suisse and Dresdner Kleinwort Wasserstein.

You think only brokerage firms and banking institutions are eagerly investigating the technology?

Dow Jones the financial news boffin is secretive about performing tests on the news-reading solution with several chief sell side firms. Meanwhile, Thomson Press corroborates- it is examining the concept with clients.

Despite the term “newsflow algorithm” akin to Algorithmic Trading, it is believed that the system will hold an immense value to the financial industry for order-generating strategies. The automated trading will consider newsflow amongst a wide array of information available for influence trading activities.

Volume and time are important factors considered by a simple algo trading. Conversely, complicated algorithms would allow several hundreds of real-time factors to arrive at a profitable investment decision. For instance, Credit Suisse swanks about an automated trading system that takes care of 3,000 data points every 10 seconds.

One-third of European and American stocks were driven by autopilot i.e., algorithms in 2006. In 2009, HFT firms accounted their 73% of US equity trading volume to the artificial intelligence.

NASDAQ, BATS and Direct Edge have gained a larger market share as compared to NYSE. Their success is attributed to algorithmic trading that enables them to reduce processing fees and commissions.

Thus, the development of automated trading has prompted a reduction in trade size and increased trade volume, leading to investment that is more profitable. And, profitable investments obtain a larger chunk of market share for automated brokerage and financial firms, as compared to old school of thought.

Algorithmic Trading Gaining Ground

Algorithmi_Trading_gaining ground

India is programmed to go the algorithmic trading way just like rest of the world since it helps to sidestep many ‘human errors’ that trip high volume traders adopting traditional trading methods.

The future of trading is in ‘programmed trading’ and India is heading the same way as other countries have been doing. Most of the day-traders lose money and it is estimated that 95 per cent of trading ended up in losses. While there is no assurance that by feeding the trading strategy into a computer, the success of it could be ensured, the chances of successful trading could go up. This is because many of the human weaknesses, like increasing the size of the order while smelling profit that would result in losses ultimately, will be eliminated and ‘that is the most important thing’.

No Human Emotion

The algorithmic trading is based on a programme written by an analyst or trader with parameters set earlier and it could be a jobbing programme or arbitrage trading etc. The advantage is that ‘human emotions will not play a part’ since the computer would execute the orders according to the programme.

It is a fully-automated process, no human intervention is needed and it is ‘away from sentiments’. The traders could do the trading in cash segments, F&O segment etc and the trading could be done in different scrips and there is the facility of putting stop loss too. The volume buying could be done at different price points giving the buyer the benefit of cost averaging and the buy/sell could be done in micro seconds. The high volatility the markets have been witnessing in the past six months has given greater opportunity for algorithmic trading.

While algorithmic trading facility is available in both NSE and BSE, because of the huge trading volume in NSE, more trades are taking place in that exchange. But one could buy in one exchange and sell in another. The algorithmic trading helped in improving liquidity and the traders could fix their own targets for buy or sell, which could be executed automatically.

Algorithmic trading is more suitable for trading in frontline stocks that had ample liquidity because when a sell order was generated, it should be executable. For HNIs, who have on hand a large volume of shares, they could set different price points for sale to average the selling price; and greater the market volatility, the greater is the opportunity in automated trading.

The Dynamics of Algorithmic Trading

Dynamics of Algorithmic Trading

Using Algorithmic Trading and High Frequency Trades could boost your trading options and perhaps your profitability too. In fact it is attracting increasing attention among market players ever since the regulator SEBI has permitted its use on exchanges .Lets analyse how technology could help you gain an edge on the markets.

Do you remember all of those slick Hollywood movies like Sneakers, Virtuosity or The Net where people use technology to gain an edge; usually monetary; over their rivals? The other day we got a peek into just such a world that at the very least left us breathless! But, at the end of the day remember that technology is a tool and it is only as good as you are!

Algorithmic Trade per se is nothing but a reflection of what happens in our brains. Algorithmic trading, also called automated trading, black-box trading, or Algo trading, is the use of electronic platforms for entering trading orders with an algorithm which executes pre-programmed trading instructions whose variables may include timing, price, or quantity of the order, or in many cases initiating the order without human intervention.

God has given us the ability , we only do this with computer procedures in a more organised or swifter manner than what God has given us . Within this broader genre exist the option of High Frequency Trades. This is basically trading that is done several times in one day, Intraday traders are High Frequency Traders. Some traders trade over a thousand times a day, and it is here that you need high speed programmes to generate your trades.

The simplified process in High Frequency Trades would start from Trade generation, through trade routing till the final step of trade execution.

You may not get the trade at the price that you expect, but you would get it at the best price in the market, this explains about the dynamics of the equity markets vis-à-vis High Frequency Trades.

Here is where the high-tech dazzle comes in. The speed of trades (in exchanges) is such is that if an exchange offers space in the exchange for your server, then you have a time advantage. Thus even the time that is taken to bounce a (trade) signal off a satellite can be avoided. While this may not be a huge issue if you were to trade say just 500 or so times a day, equations change if you trade say a million times a day.

So if there are two exchanges where there is a differential in speeds then an arbitrage opportunity exists to make money. “Arbitrage opportunities exist across time and space,”

High Frequency Trades is about volumes and not margins. Basically it is about thousands of trade and the arbitrage opportunities that lie thereof. At a personal level we feel that High Frequency Trades help in improving efficiencies in markets.

But again as said, we need to use the human mind and not technology to make money here. In High Frequency Trades you trade very fast to make money, while in Algorithmic Trading you use strategy to make money. All High Frequency Trades is Algo Trading, but not all Algo Trading is High Frequency Trades.

Algo Trading is thus, a mathematical model to trade, i.e. the timing, submission and the management of trade orders. In Dubai for example this model supports some 65% of the trading activity and in India this accounts for some 20% of trading activity in the equities arena.

And just in case one felt that the Algo model resulted in volatility in the equities segment, we must not forget that this mode of trading is just a tool and that computers and their systems just obey orders that are placed by humans. Algorithmic trading is not just a facility but and aid.

So you can use the Algorithmic system either to automatically execute a trade or as a decision support mechanism. While Algorithmic trading gives you freedom to trade, it does not replace fundamental research. It only enhances trading efficiency.

Algo Trading in India – Statistically Speaking !!!


Algorithmic trading is all the rage in India right now, and across the market the view is unanimous: the only way is up. The question is how just fast it will grow.

By the start of 2012, Algo accounted for some 24 percent of cash equities turnover in India and about 30 percent of equity derivatives. According to figures from the Bombay Stock Exchange, by far the smaller of the two Indian exchanges that dominate equities trading, the share for equity derivatives has already jumped to 45 percent since then.

Algorithms and High frequency trading are the hottest topics in the market – algorithmic trading and HFT itself, and now the regulations around it. This is what the majority of players in the market are focused on today.

India has the building blocks in place for a ramp-up. Co-location has been available from both the Bombay Stock Exchange and its bigger competitor, the National Stock Exchange, for 18 months. Both exchanges, and market observers, say their trading platforms can handle HFT. Direct market access is available. Smart order routing between the two exchanges has also been operating since August 2010.

The Indian regulator, the Securities and Exchange Board of India (SEBI), produced guidelines for algorithmic trading in March which brokers, exchanges and market watchers hail as a sensible response. The new rules, they say, recognise that algorithmic trading is a natural development and are aimed at preventing problems but not blocking growth.

“All the dynamics point to an increase in automated and algo trading in the next few years,” Expect the cash equities and derivatives levels to raise around 50-55 percent within the next year or so.

Keep up with Technology and get Your own Trading Software.

Robot Trader

Let me ask you a million dollar question, should you let a computer do the work for you, or should you trust a human to do it? I think the point is fairly obvious. So why are humans hesitant to let computers decide which stocks to trade, when to trade them, and when to take profit? A computer can monitor thousands of stocks, identify the best opportunities, and make split second buy and sell decisions faster than we can click a mouse.Computers can beat the best chess players in the world and now they are laser focused on a much bigger and more lucrative game; the stock market. If you aren’t taking advantage of this technology you probably are losing money in the market. The war has begun and it’s time to arm yourself with the best trading software.

By supporting low latency market data distribution and guaranteed messaging with a single platform and API, Greeksoft’s solutions can help your firm accelerate your trading systems from end-to-end while reducing the complexity and cost of infrastructure. Our Complex Event Processing (CEP) architecture ensures that the micro-second responsiveness remains sturdy even when market data volumes reach a high range of events per second and/or concurrent strategies stack in the thousands.

With the Greeksoft Automated Trading Software (G.A.T.S), you can not only keep up with the evolution of technology, but firmly place yourself ahead of the trading curves. So be an early adopter who recognizes a game changing technology that can give you the edge you need to be successful.

Be Smart, Go Automated.

Greeksoft Automated Trading Software : BSE Leips Algo


Greeksoft Technologies Automated Trading Software G.A.T.S is one of the leading providers for B.S.E Liquidity Enhancement Incentive Programme (LEIPS) .We have more than 120 Members trading in market making scheme through our algo’s. But what exactly is BSE LEIPS?

The Liquidity Enhancement Incentive Programme (LEIPS) was launched by the BSE to strengthen the derivatives trading platform at BSE. The programme gives cash incentives to both market makers and general market participants for trading in the SENSEX, BANKEX and 30-SENSEX stocks futures and options contracts. It is designed within the formal guidelines laid down by the SEBI in this regard. LEIPS has had a tremendous response from the investor/broker community, and more than 350 members have traded in BSE’s derivative segment after the launch of the programme.

A market maker is a member who provides for buy and sell orders within a maximum spread on continuous basis for a minimum quantity. The maximum spread and the minimum quantity can be decided by the exchange and the market maker.

Some of the significant features that characterise LEIPS are as follows:
$ It is the first of its kind liquidity-enhancement programme in the country
$ It is a transparent and all-inclusive programme
$ It offers incentives for trading and open incentives
$ It has the lowest transaction fees in the country

How to benefit from the rise of algorithmic trading .


Algo trading is an automated facility where trading is carried out by computer driven algorithms designed by traders. Instead of the traders manually doing so, it is these algorithms that determine which orders – to buy or to sell – get booked. The high speed – transactions can take as little as 18 microseconds – at which such trading takes place, gives it a competitive advantage over conventional manual trading. While a single trader can manually handle at best a portfolio of around Rs 5 crore, an algo trader, working alone, can cope with Rs 50 crore to Rs 55 crore.

Algo trading started in India in 2005. But it was only in 2008, after the Securities and Exchange Board of India (SEBI) allowed Direct Market Access, or electronic interaction with the order books of exchanges, that this facility started gaining wide acceptance. Today, around 16 to 17 per cent of trading on the Bombay Stock Exchange (BSE) and National Stock Exchange is algorithmic, with about 80 to 90 companies engaged in it. But many believe that in the next three to four years, the proportion could rise to 60 to 70 per cent.

Algo trading calls for two kinds of skills: strategy or domain knowledge, and code development. Domain knowledge means knowing stock trends in different sectors thoroughly, while code development requires a strong command of programming languages. Indeed, a background in coding is in high demand.

High Frequency Trading or HFT !!!

Algorithmic Trading

High Frequency Trading or HFT is a form of automated trading which cashes in on fleeting market opportunities through a deluge of orders executed in fractions of a second. Positions may be held only for minutes, the ‘portfolio’ is churned furiously and no position is carried overnight!

Given that they are chasing minuscule gains from high volumes, high-speed traders gain by shortening their execution time to a few milli-seconds.

Stock exchanges around the world (and in India too) have actively aided and abetted HFT by allowing market players to rent ‘co-location’ facilities at the exchange itself.

With their terminals huddled close to the exchange’s servers, co-located members — through fibre optic connectivity — execute trades at a fraction of the time that others do.