# 📊 Cheatsheet: Quantitative Trading for Beginners: Build and Backtest a Strategy

By Anh Le

> This cheatsheet is premium content. Sign in and purchase the course to view the full content.

A lookup card for the whole course: data, stats, strategy rules, backtesting, and risk. Skim the table of contents, then jump to the section you need.

## Table of Contents

| Section | Covers |
| --- | --- |
| 1. What a Strategy Needs | 1.1 |
| 2. Market Data and Cleaning | 1.2 |
| 3. Returns and Volatility | 1.3 |
| 4. From Idea to Rules | 2.1 |
| 5. Backtest Mechanics | 2.2 |
| 6. Reading Results and Overfitting | 2.3 |
| 7. Position Sizing and Risk | 3.1 |
| 8. Capstone Workflow | 3.2 |
| 9. Checkpoint Answers | 1.4, 2.4, 3.3 |

---

## 1. What a Quant Strategy Actually Needs (1.1)

A tradable strategy is a **complete, testable loop**, not just an idea.

| Component | Question it answers | Example |
| --- | --- | --- |
| Universe | What do I trade? | US large-cap equities |
| Signal | When do I want to be long/short? | 20-day momentum > 0 |
| Entry rule | Exact trigger | Close above 20-day MA |
| Exit rule | Exact trigger | Close below 20-day MA |
| Position size | How much? | 2% risk per trade |
| Costs | What does trading cost? | Commission + slippage |
| Risk limit | When do I stop? | Max 10% drawdown |

**The rule of thumb:** if you cannot write the rule so a computer could follow it with no judgment, it is not yet a strategy.

- **In-sample vs out-of-sample:** design on one slice of history, confirm on a slice you never looked at.
- **A hypothesis before a rule:** "stocks that rose over 3 months keep rising short-term" -> then translate to rules.
- **Kill criteria up front:** decide what result would make you abandon the idea.

> **Note:** 
> A strategy with no costs, no exit rule, or no sizing is a *signal study*, not a backtest. Add all three before trusting any number.

---

## 2. Market Data and Cleaning (1.2)

### Data types