About This Course

A course in technical analysis and trading-system research, taught through AmiBroker, that treats every market claim as something to be measured rather than believed.

It is published by RunBook Academy as a standalone site. RunBook Academy’s other courses cover production infrastructure; this one applies the same standard — verified against official documentation, and explicit about what was and was not checked — to a different subject.

Built with a local AI Home Lab

This website and the course content were created as a practical experiment using local LLMs running in a personal AI Home Lab.

The Home Lab is a small on-premises cluster:

• 2 × NVIDIA DGX Spark
• 1 × GMKtec EVO-X2

The models used to author the site and the course content were:

• Minimax M2.7
• QWEN3.6
• GEMMA4

The content was generated end to end against this local infrastructure. No proprietary cloud API was used to produce the material on this site.

!What has and has not been reviewed

A full human review of the generated content is still pending. What has been done is automated, and runs on every build:

• Every AFL example file and every AFL code block is checked against the official AFL Function Reference for AmiBroker 7.00.1. A function that does not exist in the reference fails the build.
• A content audit rejects unfinished sections, empty headings and missing page metadata.
• A financial-language audit rejects unsupported certainty — promises of profit, claims of prediction — in the course’s own voice.
• A link check confirms that every internal link resolves.

Automated checks catch an invented function or a broken link. They do not catch a plausible explanation that is subtly wrong. Read every lesson with that in mind: run formulas in a test database first, check statements against the official documentation for your version, and report anything that looks wrong.

What the course teaches

Technical analysis is usually taught as a set of rules that work. This course teaches it as a set of claims that can be measured. You will not be told that RSI above 70 means sell; you will be shown how to find out what actually followed RSI above 70 in your own data, how to compare that with the base rate, and how to recognise when the honest answer is “I cannot distinguish this from noise”.

The central educational philosophy: teach people not only how AmiBroker and AFL work, but how to find out whether a market claim is true.

That means every part emphasises:

• reading a chart for what it does and does not show;
• writing AFL that does what you intended, and proving that it does;
• screening, ranking and multi-timeframe analysis over a real universe of instruments;
• backtesting whose every assumption is visible in the formula rather than in a dialog;
• the mistakes — look-ahead bias, survivorship bias, curve fitting, unrealistic fills — and how to catch them in your own work;
• a research process that ends in a written report you can defend.

What is in the course

37

Parts

206

Pages

5

Levels

113

Hours

1,438

Citations

Parts	36, plus a capstone
Pages	206 lesson pages across five levels, from Chart Reader to Real-Time AmiBroker User
Hands-on	14 labs, 25 AFL projects, 7 break/fix challenges and 5 reality checks
AFL	199 example files, every one validated against the official function reference
Quizzes	895 questions, all with explanations, all working without JavaScript
Reference	129 glossary terms, 252 AFL functions taught, every source cited with a retrieval date
Validated against	AmiBroker 7.00.1

Every part is written to the same standard and none is a stub. The curriculum lists every page with its kind, its level and a time estimate derived from the content rather than chosen to sound impressive.

Before you act on anything

⚠Warning

This course is educational material. It is not financial advice, investment advice, or a recommendation to buy, sell or hold anything. Trading and investing involve risk of loss, and in leveraged instruments you can lose more than you invested. Past performance does not indicate future results.

• No strategy here is one you should trade. Several are built and tested as exercises. Each backtest is a simulation of one rule set over one sample of the past under stated assumptions — evidence about the past, not an expectation about the future.
• Run formulas in a test database first. Read the assumptions block at the top of each file and check it against your market, your instruments, your broker and your data before trusting the output.
• Check the version. Statements about AmiBroker’s behaviour are verified against the official documentation for 7.00.1 and dated. Software changes; the documentation for your version wins.
• Data-vendor claims are perishable. Provider capabilities, history depth and pricing change. Verify anything you are about to pay for with the vendor first.
• No automated trading. The course does not provide code that places orders. The chain it teaches ends at alert → human review → decision.

You are responsible for your own decisions. The full disclaimer says all of this at length.

How to read

• Read How to use this course first. It explains the page kinds, the callouts, the code labels and how to choose a starting point.
• Work through the parts in order; later parts assume earlier material. The five levels are a dependency graph, not parallel tracks.
• Do the labs, projects and challenges. They are where the understanding forms.
• Keep the glossary and the AFL function reference open while you work.
• Mark pages complete as you go; your progress shows where you are in the arc.

Verified content

Every page that makes a checkable technical claim lists its sources in front matter, with the date each was retrieved, and the citations appear at the foot of the page making the claim. Where a claim could not be confirmed against official documentation, the content says so rather than asserting it. Where the evidence for a popular market claim is weak, the course says that too.

The sources page lists every official reference used.

Platform

A fully static site built on Astro 7 and Starlight. There is no account system, no analytics, no tracking and no server-side component. Fonts are served from this site rather than from a third party, and search runs in your browser.

Your theme preference and your lesson completion are stored in this browser’s local storage — completion under the key amibroker-course:progress:v1. Clearing browser data erases the record, and there is no server-side copy. There is no export and no sync, deliberately: adding one would mean adding an account. Your progress explains the details.

Open source

Course content is licensed under CC BY-SA 4.0 and the site code under the MIT License — both, as those licences state, without warranty of any kind.

Corrections

Contributions and corrections are welcome. If you find something wrong, include the official source that contradicts the course. That is the fastest route to a fix, and it is the standard the course holds itself to.

Trademarks and independence

AmiBroker, AmiQuote, AmiFeed, AmiBroker Formula Language (AFL), AmiBroker.com and the AmiBroker logo are trademarks of Tomasz Janeczko / AmiBroker.com. Other product and company names are trademarks of their respective owners.

This is an independent educational website. It is not affiliated with, endorsed by, sponsored by or connected to AmiBroker.com, Tomasz Janeczko, or any data vendor, broker or exchange mentioned in it. No market data is redistributed here.