What is Quantitative Trading?
Quant trading has become popular in recent years. Today, however, many people don’t know what quant trading is, how it works or how to implement quant trading analysis strategies.
We want to help. In this guide, we’ll explain some of the most important things you need to know about quantitative trading so everyone can understand.
Quantitative trading is a trading strategy that involves using quantitative analysis to determine when to buy or sell. Quantitative analysis involves crunching numbers and running data through mathematical formulas.
Based on the outcome of your quantitative analysis, you might determine that a specific asset is going to rise or fall in price.
Some people call it quantitative trading, while others call it algorithmic trading.
In many cases, quantitative analysis is as simple as analyzing two of the most basic trading numbers: price and volume. In more complicated cases, quantitative analysis could require analysis of hundreds – even thousands – of different factors.
Today, some of the world’s largest investors use quantitative analysis to make informed trading decisions. A hedge fund might have a quant trading division, for example, dedicated to analyzing every trade. The hedge fund might make billion-dollar trades based on this quantitative analysis.
At a more basic level, an average investor might read quant trading analyses on the internet before making a trade. Thanks to the proliferation of quant trading guides on the internet, it’s easy for ordinary investors to implement quant trading strategies on portfolios of all sizes.
At an even more basic level, all trades involve some type of quantitative analysis. Any time you’re using math, statistics, or numbers to make a prediction about future performance, you’re engaging in quantitative analysis.
How Does Quantitative Trading Work?
The most basic quantitative analyzer involves checking two basic data inputs: price and volume. These are the two most common data inputs used in quantitative analysis.
A quantitative trading analyst might plug price and volume into a mathematical formula, for example, to make a prediction on where the asset will go next.
Picture quantitative trading as like a combination of mathematics, modern technology, and comprehensive databases. Quantitative trading throws all of these things into a blender, then extracts useful information from the resulting numbers.
Quantitative trading systems consist of four key components:
Strategy Identification: The first step is to identify a strategy. Find a strategy or create your own. Exploit an edge, then decide how frequently the system will trade.
Strategy Backtesting: Next, test that strategy on historic market conditions. How well would that strategy perform over the course of 2018? How well would it have performed in 1948?
Execution System: The next step is to link to a brokerage, automate trading, and minimize transaction costs.
Risk Management: Once the system begins executing, the goal is to optimize capital allocation and manage risk while consistently tweaking and improving the quant trading system.
Quantitative trading is a broad field. It can be combined with multiple other trading strategies. Common quantitative trading techniques can include high frequency trading, for example, or algorithmic trading and statistical arbitrage. All of these techniques rely on quantitative analysis to make informed decisions.
What Does a Quantitative Trader Do?
A quantitative trader will take a trading technique and create a model of it using mathematics. The quant trader will execute that technique using mathematical formulas.
Then, the quant trader will create a computer program that applies the model to historical market data. The model is backtested (using historical data) then optimized. If the quant trader is satisfied with the outcome, then the system is implemented in real-time markets using real capital.
In many cases, the quantitative trader uses programming languages like C++ or Python to execute these trading strategies. C++ is particularly popular for high frequency trading, although Python and R might be used for lower frequency trading.
Quant Trading is Like Meteorology: An Analogy
Our friends at Investopedia recommend thinking of quant trading like meteorology.
It’s the job of a meteorologist to analyze weather patterns, current inputs, and historical data for a specific region and then make predictions based on that information.
Just like a meteorologist, a quant trader checks various inputs, analyzes what those inputs have historically meant for markets, then makes predictions based on that analysis.
A meteorologist might release a weather report stating there’s a 90% chance of rain – even though it’s currently sunny outside. The meteorologist arrived at this counterintuitive conclusion after analyzing climate data from sensors throughout the area.
Although it’s not currently raining, historical data shows that it rains 90% of the time when similar data is detected from sensors. The sensors might have deducted a 15% drop in pressure, for example. 90% of the time when a 15% drop in pressure is detected, it’s going to rain in the next 24 hours.
A quant trader might release a similar analysis. Bitcoin’s price might be reaching $20,000, for example. The market is in full bull mode, and everyone is optimistic that prices will continue to increase. The quant trader might check underlying numbers, however, to predict that the end of the bull run is coming.
Examples of Quantitative Trading
A good quantitative trader will create a program that predicts the future.
No quantitative trading program can predict the future 100% of the time. However, a quant trading program that is right more often than it’s wrong may be able to create consistent profits.
Let’s say an investor wants to predict the future price of a stock. That investor believes in momentum investing. She writes a simple program that identifies winning stocks during an upward momentum swing in the markets. During the next market upturn, this investor’s program buys those stocks to consistently earn a profit. This is a simple example of the power of quantitative trading.
Typically, a trader will use an assortment of techniques to identify winning stocks. To complement her quantitative analysis, for example, the trader might also use technical analysis, fundamental analysis, and value investing techniques. By carefully considering all of these strategies, the trader has the best chance of picking winning stocks and maximizing returns.
Pros and Cons of Quantitative Trading
If quantitative trading was correct 100% of the time, then every hedge fund in the world would only use quantitative analysis. Quant trading, like any trading strategy, is not perfect.
Remove Emotion from Trading: Quantitative trading is all about numbers, inputs, mathematics, and formulas. A quant analysis formula has no place for emotional inputs. It’s just data.
Works Great in Conjunction with Other Trading Strategies: The best traders use a blend of strategies to inform their trading decisions. Quantitative analysis works particularly well for this purpose. It complements other trading strategies well.
Make Informed Decisions on Multiple Assets: Quant trading can quickly analyze multiple assets. Just plug the inputs into the formula to instantly get a quant analysis.
It Doesn’t Have to Be Right 100% of the Time: No trading strategy in the world is going to be 100% correct 100% of the time. But that’s not the goal with quant trading; the goal is to make more correct trades than incorrect trades.
Too Much Data: Quantitative traders have access to an enormous amount of data. You can look at market data for thousands of days of stock trading activity, for example, then develop trading strategies based on that information. Sometimes, using a lot of data is a good thing. In other cases, too much data is overwhelming for traders.
Good Quant Trading Requires Constant Adaptation: Financial markets are incredibly dynamic. A quant trading strategy needs to be equally as dynamic to keep up. A hedge fund might create an effective quantitative trading formula, only to have that formula become outdated within a few months. A quant trader might go on a winning streak when their formula is consistently delivering profits, only to go on a losing streak when their formula suddenly doesn’t work for market conditions.
You’re Competing Against Hedge Funds: Hedge funds have the money to establish a full-fledged quant trading division. They hire dozens of programmers, analysts, and statisticians to develop the best possible quantitative trading model. If you want to become a quantitative trader, you’re going to compete with these people.
How to Find or Create Quantitative Trading Strategies
Up above, we mentioned that strategy identification is the first step for implementing a quantitative trading strategy.
Finding (or creating) the right quant trading strategy today is the first step towards consistently earning profit from markets.
Fortunately, finding a good quant trading strategy isn’t hard. You can easily find profitable quant trading strategies through public sources. Academics regularly publish theoretical trading results, for example, based on various formulas and analyses. Financial industry publications and trade journals will highlight the trading strategies used by today’s leading hedge funds.
You might ask: why would someone share a profitable quantitative trading strategy? Why wouldn’t a hedge fund keep this strategy to themselves? If everybody is using a specific trading strategy, then won’t it prevent the strategy from working long-term when others crowd the market?
That’s a good question, but there’s also a good answer. Hedge funds will share the basic details of their strategies, but they won’t discuss the exact parameters and tuning methods they use to execute the trading strategy. These optimizations are crucial for turning an average strategy into a profitable one.
Here are some of the best free resources for identifying trading strategies today:
Social Science Research Network – www.ssrn.com
arXiv Quantitative Finance – arxiv.org/archive/q-fin
Seeking Alpha – www.seekingalpha.com
Elite Trader – www.elitetrader.com
These websites feature tens of thousands of trading strategies. You’ll see strategies separated into different categories, including “mean-reversion” and “trend-following” or “momentum” strategies.
You’ll also see trading strategies separated based on their frequency. Some strategies are designed for low frequency trading (LFT), for example, which typically means you hold assets for at least a day. Other strategies are built for high frequency trading (HFT), which means you buy and sell assets throughout the trading day.
You can also find “ultra high frequency trading” (UHFT) strategies, which involve holding assets for just seconds or milliseconds.
How to Backtest a Quantitative Trading Strategy
Backtesting is a crucial part of developing a quant trading strategy. After identifying your strategy, you want to see how that strategy performs on real market conditions. Fortunately, there’s a wealth of data at your fingertips, making it easy to test your strategy in historical crypto markets, stock markets, and other markets.
Many newbie quantitative traders will use the free historical trading data offered by Yahoo Finance, for example. However, more professional or advanced traders may want to pay for better data.
Free Data Versus Paid Data: Why You Should Consider Paying for Market Data
The advantage of free data is obvious: you’re getting a wealth of historical market data at your fingertips free of charge. However, there are significant downsides to free data, including:
Accuracy Issues: Free data might have errors. The data provider has no incentive to correct these errors because they’re not getting paid. Professional traders will draw data from two or more sources, then check the data against one another (say, using a spike filter) to eliminate inconsistencies.
Survivorship Bias: Many of the companies listed on stock markets in 1967 are no longer trading today. Some have been acquired. Others went bankrupt. Unfortunately, some datasets only include companies that survived the decades. This introduces survivorship bias into your strategy. You’re only analyzing companies that survived. Your trading strategy backtest will inevitably go better than it would have performed under actual market conditions.
Corporate Actions, Stock Splits, Etc.: Free datasets may also ignore certain corporate actions and how these actions affect stocks. They may not include adjustments for stock splits and dividends, for example. More professional data providers will implement adjustments into their data, but free data providers will not.
How to Setup an Execution System for your Quant Trading Strategy
Quant trading execution systems vary. Some execution systems are fully automated: the system makes trades with no manual intervention. Other execution systems are manual, with operators executing each trade.
Generally, HFT and (especially) UHFT trading strategies are fully automated, while LFT strategies are manual or semi-manual.
Some of the important things to consider when establishing an execution system include:
Interface to the Brokerage: Some people call up their broker by telephone to execute a trade. Others setup a fully-automated high-performance Application Programming Interface (API). Generally, you want your interactions with the brokerage to be automated so you can concentrate on optimizing the trading strategy.
Minimization of Transaction Costs: When making hundreds of trades in a short period of time, minimizing transaction costs is crucial. What fees does the brokerage charge? Are you paying a flat fee per trade or a percentage fee? Does the exchange charge separate fees from the brokerage? What about slippage? What’s the difference between what you intended your order to be filled at and what it was actually filled at? What about the spread? What’s the difference between the bid and ask price of the security being traded? For an average at-home investor making a few trades a month, these things don’t really matter. For quant traders – particularly HFT trades – even small fees can quickly add up.
Divergence of Strategy Performance from Backtested Performance: Some quant trading strategies work perfectly in real market conditions. They replicate their backtested success and achieve great results. Many trading systems, however, can quickly diverge, with backtested performance quickly separating itself from real world performance. Bugs can show up. Market conditions can change. The same inputs that led to certain outputs in the past may not lead to those outputs anymore.
Latency: Latency is the amount of time you lose when sending out an order. How long does it take for your order to reach the exchange or broker? Latency can significantly impact profitability – especially for HFT or UHFT strategies.
FAQs About Quantitative Trading
Q: Don’t all trading strategies involve some type of quantitative analysis?
A: By definition, quantitative analysis involves using inputs – like price and trading volume – to make predictions. Many trading strategies – even the most basic strategies – involve looking at numbers to make predictions about the future. In that sense, many trading strategies can be considered quantitative trading strategies in some sense.
Q: What’s the difference between quantitative trading and algorithmic trading?
A: Algorithmic trading and quant trading might seem like two names for the same thing. They’re closely intertwined but slightly different. Algorithmic trading is one specific part of quant trading. An algorithm developer will create an algorithm that the quant trader can use to generate profits. Without an algorithm, the quant trading developer could create a quant trading system, but it won’t work with nobody to program it. That being said, some people (including Wikipedia) use the terms quantitative trading and algorithm trading interchangeably.
Q: Which degree should I get if I want to become a quant trader or quant developer?
A: Quantitative traders come from all different backgrounds. There’s no specific degree used by the majority of quantitative traders. However, some degrees are more popular than others. Computer science and math degrees, for example, are particularly popular. Many people with a background in software development seek to enter the quantitative trading space.
Q: How can historical trading data be ‘good’ or ‘bad’. Isn’t all trading data the same?
A: Certain historical market data can be good or bad. Some data is inaccurate. Some data has survivorship bias (it only includes companies that survived to the present data). Free data sources might be good for beginner quantitative trading developers, but more serious developers will want to pay for data.
Q: Which programming language is used for quantitative trading?
A: Setting up algorithmic trading systems requires strong programming skills. Generally, C++ is the preferred language because it is the fastest, which is important when every microsecond counts. Some developers use R and Python to backtest and evaluate trading strategies, although they code in C++ for fast execution and high frequency trading. For medium and low frequency trading, any of the languages should be fine.
Q: Where I can find a trading algorithm that gives me high, risk-free returns?
A: Quant trading is advanced, but there are no guarantees of returns. If someone is trying to sell you a trading algorithm with high, risk-free returns, then you’re probably being scammed.
Palm Beach Quant + Quantitative Trading Final Word
Quantitative trading is the process of using statistics and math to predict what will happen based on what has happened in the past.
Today, everyone from cryptocurrency traders to hedge fund managers use quantitative analysis to make informed decisions. Some exclusively use quant analysis to predict their next moves, while others use quantitative analysis as part of a broader toolkit.
If you’re good at analyzing data, then you may want to get into quantitative trading. If you’re not good at analyzing data, then you can find plenty of quantitative trading resources available online where you can read quantitative analysis reports for all types of markets.
As always, happy trading!
I'm Aziz, a seasoned cryptocurrency trader who's really passionate about 2 things; #1) the awesome-revolutionary blockchain technology underlying crypto and #2) helping make bitcoin great ‘again'!