- Education
- Introduction to Trading
- What is Quantitative Trading
Quantitative Trading - Quantitative Trading Strategies
The development of computing hardware and software and increased accessibility of financial data for individuals has made quantitative trading a viable option for retail traders. While trading strategies based on extensive use of computational prowess for analyzing big data were mainly employed by big institutions in 1990s, retail traders interested in such quantitative trading strategies have also become players in the field in recent decades.
What is Quantitative Trading
Quantitative trading involves using computational methods and algorithms to develop and execute trading strategies. An example of a quantitative trading is the development of a trading platform based on a computer program that identifies winner stocks during an upward momentum in the markets, and then employing this program to buy those stocks and sell them at a profit during the next market upturn. This approach is actually a realization of momentum investing strategy in quantitative trading.
Quantitative trading techniques are employed mainly by hedge funds and high-frequency traders with short-term investment horizons who utilize near instantaneous order execution to succeed.
Quantitative Trading Strategies
Quantitative trading strategies can roughly be grouped into following types: Statistical arbitrage; Predicting the market; Market making; and Other. They are all based on distinctions in price movement predictions that are utilized to initiate deals.
Statistical arbitrage strategies aim to take advantage of relative price deviations between assets that exhibit historic relationships. The principle is to monitor prices of correlated assets and initiate deals betting on normalization of relative prices after a deviation is identified.
Strategies based on predicting the market are based on types of price predictions that draw on observations that prices exhibit tendency to move around certain levels and also exhibit certain inertia in moving in a direction. The above mentioned strategies of predicting the market can be grouped as mean reverting and Momentum strategies.
Market making strategies aim to earn profit from acting as a counterparty to both suppliers and buyers of an asset. The profit in this type of deals is made from the difference in prices a market maker pays to suppliers of an asset and charges buyers – the spread between bid and ask prices.
Other trading strategies use various data other than asset prices to help predict how asset prices will move. Examples include strategies based on monitoring statements from verified market sentiment influencers – say Warren Buffet – to trade the resulting market sentiment. Or a hedge fund may decide to buy aggregated customer spending data from financial companies to study them in order to predict a company’s performance hence the stock price.
Statistical Arbitrage
Statistical arbitrage strategies aim to take advantage of price imbalances between correlated assets on expectations that these imbalances disappear as prices revert to their normal levels. Investors often refer to statistical arbitrage as “pairs trading”.
However statistical arbitrage strategies are not limited to two securities as they invest in diverse portfolios of up to thousands of securities for a very short period of time, often only a few seconds, during the period that the price imbalance can be exploited – usually up to a couple of days.
Statistical arbitrage strategies are market neutral because they involve opening both a long position and short position simultaneously.
Mean Reversion
Mean reversion trading strategy is built upon a market’s tendency to move back to average price after an extended move. A simple version of a mean reversion strategy uses the deviation of an asset price from its average price to initiate trades – buy or sell – depending on the direction of deviation.
The average price is the average of past prices for a given number of periods. Actually, the moving average is used to determine the deviations because the average is continually recalculated based on the latest price data.
Let us say the price of a stock is trending up and thus the moving average price is also rising. We will consider the chart of closing prices for the stock. A mean reversion of the stock price will form a pattern on the price chart where the price touches or gets closer to the moving average price, then rises so that the deviation from the average price increases with successive time periods.
And then the price starts to decline so that the deviation from the average price decreases with successive time periods and price gets closer or touches the moving average price – that is the price reverts to the mean. Then the two cycles of rise and retreat repeat while the average continues rising.
In this up-trending mean reversion pattern a sell trade can be initiated when the price is high enough from the mean on expectations it will start falling and revert to the mean, and then initiate a buy trade when the price gets closer or touches the moving average. Thus a profit will be realized.
It is advisable to study various time frames and various moving averages (for different number of period say 10-period moving average and 20-period moving average, etc.) to get the feel how best to time the market entries.
Directional Strategies
Directional trading strategies follow a simple rule of initiating trades in an asset based on investor’s view of future movement of its price. Put simply: buy the asset if you expect its price to rise, and sell the asset if you expect its price to fall.
Different methods can be used to form a view on future direction of asset’s price including both fundamental analyses such as intrinsic value measurements for stocks, as well as use of technical analysis indicators to initiate long or short trades.
Directional trading is widely employed in options trading, which is associated with less risk than purchases of securities themselves and offers more flexibility.
Moving Average Crossover trading strategy is a type of directional strategy. It is based on use of technical analysis indicators - two moving averages of differing numbers of periods - to generate buy or sell orders using crossovers of those two moving averages.
When charting two moving averages of differing numbers of periods, say price moving averages for 10 and 20 periods denoted as MA(10) and MA(20), the average of shorter term (relative to the other moving average) – the MA(10), will move in the same direction as the longer term moving average MA(20) - but at a higher rate.
The different rates of change result in points where the values of the two moving averages may equal and or cross one another. These points are called the crossover points.
These crossovers are used as points of decision to generate buy or sell orders. When the shorter term moving average MA(10) moves above the MA(20) , that is used as a Buy or (Long) signal. The logic behind it is that the shorter term moving average’s move above the longer term moving average is interpreted as a lagged indicator that the price is moving upward relative to the historical price. Similarly, a Sell (or Short) signal is generated when the shorter term moving average moves below the longer term moving average.
Market Making
Market making is known as a market neutral trading strategy used for securities traded on exchanges. Market making is a trading strategy where a trader provides limit orders on both sides of a mid-price. A limit order is an order to buy or sell a security at a specific price.
Traders employing market making strategy earn their income on the bid-ask spread and trading volumes. As this process increases the liquidity in the market, it is known as market making.
As a market maker provides liquidity to the order book of a certain asset, he constantly updates the price based on the supply and demand in the market.
Market making strategies are necessarily based on high frequency data. They demand quantitative models to identify clusters of persistent market conditions (known as market regimes) that affect the success of trading strategies.
Algorithms designed for market making strategies are fed with tick level transaction data and a real time copy of the order book of the exchange.
Quantitative Trading vs Algorithmic Trading
Quantitative trading refers to development of trading strategies that are based on advanced mathematical models. It involves using complex mathematical and statistical models to analyze historical data and exploit trading opportunities in order to make a profit.
In algorithmic trading, quantitative models are employed to determine important parameters of a trade such as the price, timing and quantity of an asset, as well as execute the trade automatically without human intervention.
Algorithmic trading makes use of a pre-programmed algorithm to automate the full process of trading - including order generation, submission, and the order execution. Algorithmic trading thus is a subset of quantitative trading.
While both quantitative trading and algorithmic trading are used mainly by financial institutions and hedge funds, once the trading strategy is built in quantitative trading the trades can be executed both manually or automatically using those strategies.
Pros and Cons of Quantitative Trading
One of the primary advantages of quantitative trading is it eliminates human intervention in trading. After the strategy is developed and the system is built based on that strategy, trading is done hands-off by computer, eliminating human behavioral mistakes. This allows the trader to follow the trading plan and not depend on personal discipline in implementing the plan for successful trading.
Another primary advantage of quantitative trading is it requires little time after the strategy is incorporated into the system. There is no need to sit in front of the monitor all day, and the freed time can be spent on getting additional income.
Further advantage of quantitative trading is it makes it possible to trade enormous number of strategies and portfolios. This allows to spread capital across different markets and time frames, making for more diversified trading.
The main constraint of quantitative trading is that it requires a far larger capital commitment than usual retail discretionary trading. And in order to actually engage in quantitative trading one has to be able to build automated trading system by doing certain amount of programing.
So, one has to have a certain level of programing knowledge which is necessary for implementing as well as for back-testing the trading strategies he develops. In case of absence of such knowledge one must rely on hired help to back-test and build the trading system.
Quantitative trading requires also higher quality price data, live data feeds, a virtual private server, also known as a VPS, and software. These may cost up to a few hundred US dollars a month, depending on level of service.
Bottom Line on Quantitative Trading
Quantitative trading comprises wide range of trading strategies that can be implemented on automated trading platforms based on mathematical analysis of historic data. It requires higher capital investment compared with usual retail trading but offers opportunity to select effective strategies based on back-testing on historic data.
