Beginner’s Guide to Quantitative Trading I: Useful skills and where to find them
Auquan recently organized QuantQuest — a trading hackathon to find the best quant trader. During the course of this competition, which saw 1400+ registrations, we received overwhelming requests for a guide on how to get started with quantitative trading. Many of our participants were talented, experienced professionals working as full time developers or data scientists who wanted to make forays into trading.
The internet is a wonderful place, with tons of resources on how to develop and hone your trading abilities — but that is it’s curse too. With so much help available, where do you begin? How do you progress? What knowledge is necessary to make novice attempts at trading and what skills should you supplement your beginner learnings with to take your trading to next level?
Before you despair, we’re here to help. Here we attempt to lay down a rough guide for you with links to online resources to get you started on your path to be a star trader. You may find that you are already proficient in some of these skills, while other may require more serious work.
In a previous post, we mentioned the key to successful mastery of quantitative trading is getting the math right and backing it up with functional knowledge of a statistical programming language like Python/R. Then start applying these skills to some simple trading strategies. If you can pick up some data science skills along the way, even better!
So let’s get started(the topics are listed in order of how you should be building up your skillset, however too much of the same is always boring. To keep things interesting, you can try working on a couple of things at the same time):
1. Math and Statistics:
I cannot stress how important this is. Statistics is the foundation of quantitative trading, most of the work is getting math of the data right. We have a basic math refresher series for you:
- Random Variables and distributions
- Expected Value
- Correlation and Confidence Intervals
- Cointegration and Stationarity
If this piques your interest, you can follow it up with our tutorials on basic time series analysis.
For those who still prefer their learnings to come from textbooks, Introduction to Probability Models by Sheldon Ross is the accepted standard for introduction to probability and stats, and is my go to anytime I need a quick revision.
2. Programming:
It is recommended that you build a functional knowledge of a statistical programming language like Python or R. Our toolbox is built with Python, so we list a few useful resources below:
- Code Mentor has a quick starter Cheat Sheet and Data Camp has a quick Intro to Python course.
- Our toolbox extensively uses Pandas. Here’s a cheat sheet and a quick tutorial.
- For more detailed descriptions, check out learnpython.org
- If you have little programming experience and are looking for a comprehensive course, Code Academy has a good offering.
- DataQuest has a great platform that I often use for quick data manipulation hacks.
3. Simple Trading Strategies
Once you’ve built-up your programming and math skills, it’s time to get to real work. We provide a beginners’ toolbox to get a flavor of working with financial data.
- Check out our series on beginner trading strategies(with iPython Notebooks) that talks about model selection, mean reversion, momentum trading and other simple strategies
- Quantstart has a good guide to mean reversion strategy
- 101 Formulaic Alpha provides you some ideas for inspiration
- And here’s another with basic to advanced strategies
4. ML — the new kid (ok maybe not that new) on the block
Machine Learning is all the rage these days. And it does deserve the hype it’s receiving. So if you want to be a rockstar quant, better get some of these cool tools in your kit.
- Scikit is the best python package to get started with ML. For a tutorial, check out Data Camp’s tutorial.
- Find a quick ML tutorial here and a crash course here.
- Build Better Strategies provides a quick tutorial on how to develop a ML based trading strategy from scratch .
- David Aronson’s experiments with ML for financial predictions also form a good read.
5. Successful Backtesting and Metrics
Building profitable strategies is not just about finding an amazing idea or new signal, your strategy needs to be tested and optimized and verified rigorously before being put into production. We have some resources that talk about backtesting best practices and how to evaluate trading strategies.
- Read our post about trading metrics and evaluating trading strategies
- We also have resources on backtesting biases and risk management.
- Read more about backtesting biases
6. Other Reading
Developing profitable strategies requires work and constant efforts. The resources above will help you get started with the process, and then the onus is on you to continue to supplement your skills and knowledge. Below is a list of some of the blogs we regularly follow to stay updated:
This post is mostly meant to list useful resources to get started with writing trading strategies. For a step by step guide on how to develop an automated trading system, read on to part II.
