Chapter 1: Introduction to Quantitative Trading¶

Welcome to the world of systematic trading.

What is Quantitative Trading?¶

Quantitative trading uses mathematical models and computer programs to identify and execute trading opportunities:

Text Only

Data → Analysis → Signal → Decision → Execution

Instead of relying on intuition or gut feelings, quant traders:

Analyze historical data
Build mathematical models
Test strategies rigorously
Execute systematically

Systematic vs. Discretionary¶

Aspect	Discretionary	Systematic
Decision Making	Human judgment	Rules-based
Emotions	Can interfere	Removed
Consistency	Variable	Consistent
Scalability	Limited	High
Speed	Slow	Fast
Backtesting	Difficult	Easy

Types of Quantitative Strategies¶

1. Statistical Arbitrage¶

Exploiting price relationships between related securities:

Text Only

signal pairs_trade:
  // Trade the spread between related stocks
  spread = prices[GOOGL] - beta * prices[META]
  zscore = (spread - mean(spread)) / std(spread)
  emit -zscore

2. Factor Investing¶

Systematic exposure to return drivers:

Text Only

signal multi_factor:
  momentum = zscore(ret(prices, 60))
  value = zscore(book_to_market)
  quality = zscore(roe)
  emit 0.4 * momentum + 0.3 * value + 0.3 * quality

3. Trend Following¶

Riding market trends:

Text Only

signal trend:
  ma_fast = rolling_mean(prices, 20)
  ma_slow = rolling_mean(prices, 60)
  emit zscore(ma_fast - ma_slow)

4. Mean Reversion¶

Betting on prices returning to normal:

Text Only

signal mean_reversion:
  z = (prices - rolling_mean(prices, 20)) / rolling_std(prices, 20)
  emit -z  // Buy oversold, short overbought

5. Market Making¶

Providing liquidity for a profit:

Text Only

Not covered in this guide (requires low-latency infrastructure)

The Quantitative Trading Process¶

1. Research¶

Generate ideas from academic papers, market observations
Formalize hypotheses mathematically
Initial data exploration

2. Signal Development¶

Text Only

// Transform ideas into computable signals
signal my_idea:
  // Mathematical implementation
  emit zscore(my_calculation)

3. Backtesting¶

Text Only

// Test on historical data
portfolio test:
  weights = rank(my_idea).long_short(top=0.2, bottom=0.2)
  backtest from 2015-01-01 to 2024-12-31

4. Validation¶

Out-of-sample testing
Walk-forward analysis
Parameter stability

5. Production¶

Deploy to live trading
Monitor performance
Manage risk

Why Use sigc?¶

sigc provides a complete platform for quantitative trading:

Domain-Specific Language¶

Write strategies naturally:

Text Only

signal momentum:
  emit zscore(ret(prices, 60))

Rigorous Backtesting¶

Test with realistic assumptions:

Text Only

portfolio main:
  weights = ...
  costs = tc.bps(10)  // Transaction costs
  backtest from 2015-01-01 to 2024-12-31

Production Ready¶

Deploy the same code to live trading:

Bash

sigc daemon start --strategy momentum.sig --live

Performance¶

Built in Rust for speed: - Process millions of data points - Run parameter optimizations - Execute in real-time

Key Concepts Preview¶

Signals¶

Numerical scores indicating expected returns:

Text Only

signal alpha:
  // Higher score = expect higher return
  emit zscore(ret(prices, 60))

Portfolios¶

Converting signals to positions:

Text Only

portfolio main:
  weights = rank(alpha).long_short(top=0.2, bottom=0.2)

Backtesting¶

Simulating historical performance:

Text Only

backtest from 2015-01-01 to 2024-12-31

Risk Management¶

Controlling exposure:

Text Only

constraints:
  max_position = 0.03
  max_sector = 0.20
  net_exposure = 0.0

Who Should Use This Guide?¶

Aspiring Quants¶

Learn the fundamentals of systematic trading from scratch.

Software Developers¶

Understand financial concepts and sigc's approach.

Finance Professionals¶

Transition from Excel to programmatic trading.

Data Scientists¶

Apply ML/statistics skills to financial markets.

What You'll Learn¶

By the end of this guide:

Understand quant trading fundamentals
Build trading signals from data
Backtest strategies properly
Manage portfolio risk
Deploy to production
Monitor live trading

Prerequisites¶

Minimal Requirements¶

Basic programming concepts
High school mathematics
Interest in financial markets

No Need For¶

Prior Rust knowledge
Advanced mathematics
Professional trading experience

Getting Started¶

Install sigc¶

Bash

cargo install sigc

Your First Strategy¶

Text Only

// Save as first_strategy.sig
data:
  source = "prices.csv"
  format = csv
  columns:
    date: Date
    ticker: Symbol
    close: Numeric as prices

signal momentum:
  emit zscore(ret(prices, 60))

portfolio main:
  weights = rank(momentum).long_short(top=0.2, bottom=0.2)
  backtest from 2020-01-01 to 2024-12-31

Bash

sigc run first_strategy.sig

Next Chapter¶

Continue to Chapter 2: Fundamentals to learn about returns, volatility, and other core concepts.