Real life investment scenarios, may it be personal investment decisions or business decisions are more complex than what we learn in classes. You can’t afford to make mistakes because of the shear scale of these decisions. The use of models can be of a great help in such situations. Models help us picturise the uncertainty in future and act accordingly in present. Anything which can be stated or measured in number can be quantified. The quantified information coupled with mathematical and statistical modelling gives rise to something known as quantitative models. There are two types of analyses, qualitative and quantitative. Qualitative analysis depends on meaning of the scenario and focuses more on the sensitivity of the context.
Quantitative models are used more in financial sector institutions. They make use of complex mathematical and statistical formulae to spot investment opportunities having small horizons. In many cases the margin is smaller but the funds in use are huge making it viable to use these techniques. High frequency trading, algorithmic trading uses quantitative models in equity markets. High speed computers doing millions of transactions in small time frame and decision making in milliseconds without nearly no human intervention is the speciality of using quant models in high frequency trading. They make use of the price and volume data to make decisions.
Arbitrage opportunities are also exploited by the algorithms in place. As these techniques need high speed computers, institutional investors are benefited at the expense of small pocket investors. The idea has met criticism in the past because of various disadvantages it brings with it. The high liquidity it creates can disappear in seconds as it is brought in for transactions which are executed in matter of milliseconds. The algorithmic trading is gaining popularity among small investors as well. Companies are trying to use amateur traders or code writers to write better formulae for them in exchange of commission. In 2006 the share of algorithmic trading was as low as 25% and within ten years it went on to more than 80%. The question at hand is whether these models and techniques can be used for cryptocurrency trading.
What do you mean by currency trading?
When we have a forex trading account, we decide on a pair of currencies which we want to trade. Buy or sell one currency in exchange of another depending on our interpretation of the current market scenario and expectations in the rise or fall of the currency value. In simple words cryptocurrency trading is also a kind of currency trading with one currency digital and other the fiat one or both digital. There can be numerous pairs of currencies like Bitcoin-USD where one buys or sells the cryptocurrency and gains from the price movement.
Cryptocurrency trading is a bit different in many ways. The markets are open 24/7 unlike traditional currency markets. It is still at nascent stage because of which it is not saturated by big bulls using algorithmic trading. The most important thing about cryptocurrency market is the volatility. These markets move up and down in a big way because of numerous things including regulations, perceptions, technological advancements etc. It can be understood from the fact that, the variation in BTC-USD pair in last 2 years has been from as low as $3000 per bitcoin to $10000 per bitcoin. On the other hand, the variation in EUR-USD pair has been quite nothing in that comparison. Source: Yahoo Finance
BTC-USD Chart
Arbitrage: The best example of arbitrage is when there is price difference for the same security in different markets or exchanges. You can buy at low price on one exchange and sell at higher price on another exchange. It may sound easy on paper but can be a big task when it comes to execution. A faster and smarter quant model can be of a great help.
Market Making: There are many players taking part in the market. These are brokers, buyers, sellers, market makers etc. Brokers just play a role in connecting buyers with sellers. Market makers generally keep securities with them thus helping the buyers to buy securities as and when they want it. They are rewarded for the risk of keeping securities. There can be times when the price of securities goes down after the market maker buying it from the seller. Generally, they try to earn on the spread between the bid and the ask price.
Iceberg Technique: Few orders can be so huge that they have the potential to halt the market or change the course of actions to the least. In such situations, it is better not to reveal the complete order to the market thus preventing other users from acting in reverse direction. The complete order is thus divided in smaller orders which doesn’t lead to a heavy price fluctuation.
These are very few of many strategies which are used in algorithmic trading. There can be simple formulae used for spotting opportunities and there can be difficult statistical techniques as well. To the core of everything is the need of being the first to get there and exploiting the chance to earn profit. Quantitative models are still not as popular as they are in traditional markets. There are great challenges and opportunities in this area which still looks unexplored.
_____________________________________________________________________________________
About the author: Vaibhav Thakare is an MBA student from IIM Lucknow. An engineer who has worked in Strategy and Trade Finance, he has always been interested in learning about new technologies and innovative solutions. Writing is something which excites him because it’s about helping someone understand these difficult looking topics in simple words.
