Understanding — Algorithmic Trading
Gone are those days when on imagining any typical day at the Wall Street trading offices, we are immediately left with an image of a large screen displaying multiple small red & green pyramids & a bunch of hyper tensed brokers shouting at the top of their voices to quote price after price of various stocks, with their every passing breathe. Now the same allies have turned much quitter with the trades being registered by quietly humming, yet powerful computers.
The evolution of computers from being our digital slave to humans for performing repetitive and mundane tasks or calculations, to fully replacing humans from a high-intensity decision-making profession is quite remarkable. These computers (as NextGen brokers) have provided the retail investors with the flexibility, to execute the trades while being at any place in the world.
But to further push these boundaries we started with experimenting on providing complete autonomy to our emotionless brokers -computers. Hence the new way of trading emerged — Algorithmic Trading.
Algorithmic Trading
Algorithmic (or Algo) trading is a trading method where we define a set of parameters in a form of codes to the computers which results in the output as a command, on whether to buy a stock or sell a stock. The detailed stages of Algo trading are shown in the image below:
Stages of Algorithm based Trading
Pre-trade Analysis:
This stage particularly deals with a range of stages from — Integrating data from various terminals to a central company database to provide the various parameters to the trading algorithm, Selection of Asset class to trade for the particular day based on the volatility in the primary & secondary market; to specify the expected risk in terms of the draw-on that could be faced by the system upon deploying the algorithm in the live market.
Trading signal generation:
In this stage, the trading algorithm will generate various signals based on the pre-specified indicators or parameters. A few of the commonly used indicators are — the volatility models for assets class markets (Alpha Modeling), Position sizing to optimize the risk-reward ratio on a given trading day(Risk Modelling), & Optimization of the number of trades based on the market making provided by partnering broker of the firm (Transaction Cost Model).
Trade execution:
There is 2 way to utilize the algorithm-based trading in execution stage — either let the algorithm fully handle the trade decisions based on the pre-determined set of rules, or let the algorithm generate the trade signals for the trader sitting on the trading deck then he will pick & choose the best possible trade setup that he wants to trade.
Further, another crucial aspect of trade execution is execution speed. Often the fully governed automated algorithm-generated high-frequency traders (HFTs). Thus, the prop desk also has to decide whether to take a trading platform service from an established broker or have an in-house trade platform extension directly connecting to the stock exchange provider.
Post-trade analysis:
This stage broadly covers the report generation of the trades that happened daily & fulling the compliance requirements of the regulators. The daily trade reports are very crucial for any prop desk provider because by thoroughly analyzing daily trades they can optimize their strategies based on the changing market conditions.
Pros & Cons of Algo Trading
Pros:
- Eliminates discretionary trading. Thereby reducing the dependency of the skills of trades to generate constant returns
- Automates the data or insights collection from various sources, which are crucial for trades
- The bulk of orders could be executed directly at very high frequencies
- Risk Management could be factored indirectly in the algorithms, then by reducing the human errors
Cons:
- Algorithms need to be updated frequently to remain relevant for changing market conditions
- Even with fully automated algorithms, human monitoring is required to ensure that the system is behaving in the desired manners
- The cost of building an algorithm design team is often high, as it requires a good combination of veteran trades & skilled programmers
Conclusion: Though technological advancements have transformed the working of Stock Markets by providing various functionalities to retail investors, it has also raised various red flags on the extent to which Stock Markets could be manipulated. Hence in my next article, I will shed some light on the market manipulations.
