Quantitative trading for beginners: what to learn, which tools, and in what order
A beginner path. What to learn first, what comes next, and which page to read.
Short answer: what to learn first: a little Python and pandas, a China A-share history that updates itself, one backtest, how to read the result and find the traps, then paper trading, and only then a live account. Quant trading does not guarantee a profit. Stopping at the backtest or the paper account is normal, and it is cheaper than finding out with real money.
How to choose: Follow this order. Learn a little Python and pandas. Keep a China A-share daily history that updates itself. Run one backtest. Learn to read the result and to find the usual traps. Run a paper-trading period. Only then decide whether a live account is worth it. Quant trading does not guarantee a profit. Many people stop at the backtest or the paper account. That is a normal outcome, and it is cheaper than finding out with real money. One data library and one backtester are enough at the start, for example AKShare and backtrader. This page is the order of what to learn, with one site guide at each step, plus a short book list. It does not set the order of tools. It is not the site FAQ. It does not publish a broker threshold or a fee table. Those follow the broker's own notice.
Who it is for: Someone new to quantitative trading who wants a sequence: what to learn first, which guide to read at each step, and which books have a free online edition or a checked Chinese edition. A little Python is enough to start, or a willingness to learn it. If you do not code yet, hosted sites such as JoinQuant, RiceQuant, and BigQuant can be a first look, and so can an AI stock-picking assistant or a TradingView chart. Owning the whole loop still means coming back to this path and learning some Python.
Who it is not for: Someone who wants a promised return, or a broker's account minimum and fee table on this page. Thresholds and fees follow the broker and each platform's own notice. The platform comparison is a different article. Someone who already has a backtest and only wants the paper-trading ledger should read that walkthrough. Someone who wants the separate English study plan should read that page. It is not this article. This page does not set the tool order and it is not the site FAQ.
The same guide in Chinese is the beginner learning path. A separate English study plan, which is not this article, is what to learn, in what order. None of the step guides below is a language version of this page. This page is not the tool-order guide and not the site FAQ.
This site has no relationship with the publishers, platforms, or brokers named below. There are no purchase links. Book editions and Chinese-edition details were checked on 22 Oct 2026 against author sites, publisher pages, and Douban listings. Anything that check did not settle is left out. This is not investment advice.
01 · What quant trading is, and what it does not promise
Quantitative trading writes “what to buy, when, and how much” as an explicit rule, tests that rule on history, and only then lets a program follow it. It answers whether the rule was applied consistently and whether history has been asked. It does not answer whether the rule itself will make money.
- Difficulty. Starting is not hard. A little Python is enough for a first backtest. The hard part is the checking that follows: whether the data is right, whether a future fact slipped in, whether costs were included, and whether another period still agrees.
- Profit. Not guaranteed. A profitable backtest and a losing live account is the usual case, not the exception. Overfitting, fees and slippage left out, and a market that changed are the usual reasons.
- Why learn it anyway. Even if you never send a live order, the loop turns a vague trading idea into a rule you can check against history.
02 · What you need before you start
- Python. Variables, lists, dictionaries, loops, functions, and the ability to read someone else's example.
- pandas. Price data is a table. You will read it, filter it, align it by date, and take rolling averages. The first book below is about that.
- A little statistics. A mean, volatility as a standard deviation, and correlation. You need them when you read a backtest. You do not need a course before you start. Fill the gap when you meet it.
You can try the work before you can program. Hosted sites such as JoinQuant, RiceQuant, and BigQuant already have data and examples. That comparison is how to choose a quant trading platform. If you only want AI to help pick stocks, start with how to choose AI stock-picking tools. If you want to learn charting and a screener first, read the TradingView setup.
03 · Six steps: what to learn, and what “done” means
| Step | Learn | Done when | Read |
|---|---|---|---|
| Step 1 · Foundations | Python basics, and pandas: read a table, filter it, and align rows by date. Have a Python environment of your own. | You can load a daily price table, compute each day's return, and compute a 20-day average. | Local Python and SQLite (Chinese); A cloud quant desk |
| Step 2 · Get data | What each free data source gets wrong, and how to store a history that updates every day. | You have a local China A-share daily history that updates itself. | How to choose a market data source; Tushare and AKShare; AKShare into SQLite |
| Step 3 · One backtest | Pick one backtesting framework and run one simple strategy, such as a dual moving-average cross. | You can say which period you tested, which costs you included, and what the result was. | backtrader: a first strategy; How to choose a backtesting framework |
| Step 4 · Read the result | How to read return, maximum drawdown, and Sharpe, and the three usual traps: lookahead, overfitting, and costs left out. | You can point at a flattering backtest and say where it may be too good to be true. | How the backtrader walkthrough reads a result |
| Step 5 · Paper trade | Run for a while on live quotes, with no real money, and compare that with the backtest. | The paper period has run long enough, and the gap is one you can accept. | Paper trading after a backtest |
| Step 6 · Only then, a live account | How a broker quant terminal is opened, where the code runs, and that an individual reports programmatic trading first. | You know the opening conditions, the fees, and the reporting rule, and only then decide whether to use real money. | How to choose a quant trading platform; QMT and PTrade |
A few notes:
- Few tools first. One data source and one backtesting framework are enough at the start, for example AKShare plus backtrader. The backtrader repository has gone a long time without an update. It is still a reasonable way to learn the ideas. To switch frameworks, use the framework comparison.
- Step 4 is the one people skip. When a backtest looks good, doubt it before you celebrate. The most common mistake is deciding today's trade from data that was known only after today's close. That is lookahead.
- Read the rule before step 6. A program that connects to your account and sends orders is programmatic trading. Under the Shanghai Stock Exchange notice, programmatic stock trading on the SSE is reported to the broker before the first time. The detail is section 07 of QMT and PTrade.
04 · What it costs to start
- Research and a backtest. Can cost nothing. AKShare, backtrader, and Qlib are free and open source and run on your own computer.
- Data. AKShare is free and can fail without notice. Tushare's free tier can pull unadjusted daily bars only. Adjusted prices and financials need a paid points tier. The published prices are on Tushare and AKShare. This page does not restate them.
- A server. Optional. You need one only if a script should run every day while your own computer is off. Opening the machine and connecting over SSH is the cloud server setup.
- A live account. Broker quant terminals differ in thresholds and fees. Follow the broker you open the account with. How far each platform goes without paying is section 06 of how to choose a quant trading platform.
05 · After the six steps: three next directions
- Machine-learning stock selection. Start with the Qlib walkthrough. That page records a run on Windows.
- A language model for research. Deploy an open-source research agent. The Docker walkthrough is one-click Docker deploy. The same walkthrough in Chinese is one-click Docker deploy. More projects are in open-source AI quant projects. The same list in Chinese is the Chinese list of AI quant projects.
- Trading bots and crypto. The rules and the risks differ from China A-shares. Start with how to choose a trading bot.
06 · Books
Only books whose author and edition were checked. Chinese-edition details come from Douban listings. Confirm them with the publisher before you buy. No purchase links.
- Read this first. Wes McKinney, the author of pandas, Python for Data Analysis, 3rd edition. Published August 2022, updated for pandas 2.0 and Python 3.10. The author site hosts a free online edition. The Chinese edition is China Machine Press, October 2023. The Chinese title is on the Chinese page. It does not teach quant trading. It teaches the tables that quant work uses every day, and it is the cheapest place for a beginner to start.
- After a first backtest. Stefan Jansen, Machine Learning for Trading, currently the 3rd edition. The companion code is public on the author's GitHub repository under the MIT licence, and it is listed on this site at Machine Learning for Trading. The 2nd-edition code is on the repository's second-edition branch. A Chinese edition was not found. A publisher and a year for the 3rd edition were not confirmed, so neither is stated here.
- When you want the machine-learning failure modes. Marcos López de Prado, Advances in Financial Machine Learning, Wiley, 2018. It is about the mistakes that are easy to make with machine learning on financial data. It is math-heavy, and it is a poor first book. The Chinese edition is CITIC Press Group, 2021. The Chinese title is on the Chinese page.
- Optional. Yves Hilpisch, Python for Finance, 2nd edition, O'Reilly. It covers a lot of ground. The author's code repository says the code is based on Python 3.7, so typing along means watching for version gaps. A Chinese translation of this edition was not found.
Related English guides: a first backtrader strategy · paper trading · quant trading platforms · a separate study plan · all guides · all tools
Frequently asked questions
How to choose a quant learning path?
Follow this order. Learn a little Python and pandas. Keep a China A-share daily history that updates itself. Run one backtest. Learn to read the result and to find the usual traps. Run a paper-trading period. Only then decide whether a live account is worth it. Quant trading does not guarantee a profit. Many people stop at the backtest or the paper account. That is a normal outcome, and it is cheaper than finding out with real money. One data library and one backtester are enough at the start, for example AKShare and backtrader. This page is the order of what to learn, with one site guide at each step, plus a short book list. It does not set the order of tools. It is not the site FAQ. It does not publish a broker threshold or a fee table. Those follow the broker's own notice.
Who it is for?
Someone new to quantitative trading who wants a sequence: what to learn first, which guide to read at each step, and which books have a free online edition or a checked Chinese edition. A little Python is enough to start, or a willingness to learn it. If you do not code yet, hosted sites such as JoinQuant, RiceQuant, and BigQuant can be a first look, and so can an AI stock-picking assistant or a TradingView chart. Owning the whole loop still means coming back to this path and learning some Python.
Who it is not for?
Someone who wants a promised return, or a broker's account minimum and fee table on this page. Thresholds and fees follow the broker and each platform's own notice. The platform comparison is a different article. Someone who already has a backtest and only wants the paper-trading ledger should read that walkthrough. Someone who wants the separate English study plan should read that page. It is not this article. This page does not set the tool order and it is not the site FAQ.
Is quantitative trading hard?
Starting is not hard. Doing it well is. With a little Python, a day or two of following a walkthrough is enough to run a first backtest. The hard part comes after that: whether the data is wrong, whether the backtest peeked at the future, whether costs were included, and whether the rule still works on another period. Most of the time goes into those checks, not into writing the strategy.
Can quantitative trading make money?
It does not guarantee a profit. Quant trading writes a rule as a program and tests it on history. The program does not decide whether the rule is any good. A backtest that makes money and a live account that loses it is common. Usual reasons are overfitting, fees and slippage left out, or a market that changed. Backtest, then paper trade, and consider real money only if the gap is one you can accept.
Can I do quant trading without programming?
You can do part of it. Hosted sites such as JoinQuant, RiceQuant, and BigQuant already have data and examples, and BigQuant also has a visual, drag-and-drop style. If you do not want to write code and only want AI to help pick stocks, start with the AI stock-picking guide. To own the whole loop yourself, you still have to learn some Python.
How much does it cost to start?
Research and a backtest can cost nothing. AKShare, backtrader, and Qlib are free and open source and run on your own computer. What can cost money is more stable data, such as a paid Tushare points tier, an optional cloud server, and whatever threshold a broker's quant terminal may have. Those follow each vendor's own notice. This page does not restate a price.
What book should a beginner read?
Start with Wes McKinney's Python for Data Analysis, 3rd edition. The author site hosts a free online edition. The Chinese edition is China Machine Press, October 2023. The Chinese title is on the Chinese page. After a first backtest, read Stefan Jansen's Machine Learning for Trading and Marcos López de Prado's Advances in Financial Machine Learning. The Chinese edition of López de Prado is CITIC Press Group, 2021. The Chinese title is on the Chinese page.
Do individuals have to report programmatic trading to the exchange?
If a program connects to your account and sends orders, that is programmatic trading. Under the Shanghai Stock Exchange notice, an investor who conducts programmatic stock trading on the SSE reports to the broker before doing so for the first time. Research, a backtest, and paper trading are not that step. The detail is in the QMT versus PTrade guide, section 07.
This is not investment advice. Quant trading can lose money. A backtest and a paper account are not a forecast. Books, platforms, and brokers follow their own current notices.