Quant Trading System

A multi-factor quantitative trading system for medium-term US equities. Combines six alpha factors with mean-variance portfolio optimization and automated execution via Alpaca.

Strategy Overview

The system scores stocks using a weighted composite of six factors:

Factor	Weight	Description
Momentum	30%	Cross-sectional returns over 1/3/6/12 months (skipping most recent month)
Trend	20%	SMA 50/200 ratio — favors stocks in uptrends
Value	15%	Inverse P/E and P/B ratios — favors cheaper stocks
Mean Reversion	15%	Bollinger band z-score — favors oversold names
Volatility	10%	Realized 63-day vol — favors low-volatility stocks
Quality	10%	ROE, profit margins, earnings growth

The portfolio holds up to 20 positions, rebalances monthly, and targets 15% annualized volatility.

Quick Start

Install

pip install -r requirements.txt

Dependencies: numpy, pandas, yfinance, scipy, matplotlib, seaborn, ta

Run a Backtest

# Default backtest (2022-01-01 to today, $1M capital)
python run.py backtest

# Custom dates with plot output
python run.py backtest --start 2023-01-01 --end 2024-12-31 --plot

# Verbose logging
python run.py -v backtest --plot --plot-output my_backtest.png

View Current Signals

python run.py signal

Shows the composite alpha score for every stock in the universe, ranked from strongest to weakest.

---

Manual Usage (Google Colab)

The notebook quant_trading_colab.ipynb gives you a visual, interactive workflow — no local setup needed.

Setup in Colab

1. Open the notebook in Google Colab
2. Run Cell 1 (Setup) — clones the repo and installs dependencies
3. Run Cell 2 (Imports) — loads config and initializes the strategy

What Each Cell Does

Cell	Purpose
1 - Setup	Clone repo, install packages
2 - Imports	Load config, create strategy object
3 - Backtest	Run full historical backtest, print summary metrics (Sharpe, CAGR, max drawdown, etc.)
4 - Equity Curve	Plot equity curve, drawdown, and rolling Sharpe ratio
5 - Monthly Returns	Heatmap of monthly returns by year
6 - Risk Report	Extended stats: skewness, kurtosis, VaR, CVaR, win rate
7 - Current Signals	Fetch today's alpha scores for all stocks
8 - Target Portfolio	The key cell — shows exactly what to buy today

Using the Target Portfolio (Cell 8)

Edit MY_CAPITAL to match your account size:

MY_CAPITAL = 50_000  # <-- your portfolio size in USD

Run the cell. Output looks like:

=================================================================
  TARGET PORTFOLIO  |  Capital: $50,000
=================================================================
  Stock      Weight    Dollars   Shares      Price     Score
  ------------------------------------------------------------
  NVDA        9.8%    $4,900      36   $136.25    0.842
  META        8.5%    $4,250       7   $607.14    0.731
  ...
  ------------------------------------------------------------
  TOTAL      95.2%   $47,600
  Cash reserve: $2,400
=================================================================

Then manually place the trades in your brokerage account.

Rebalancing Manually

Run Cell 8 monthly (every ~21 trading days). Compare the new target with your current holdings and adjust:

• Sell positions that dropped out of the top 20
• Buy new positions that entered
• Resize existing positions to match target weights

---

Automated Trading with Alpaca

1. Create an Alpaca Account

Sign up at alpaca.markets. Start with a paper trading account to test without real money.

2. Set Environment Variables

export ALPACA_API_KEY="your-api-key"
export ALPACA_SECRET_KEY="your-secret-key"

Paper trading is the default. For live trading (at your own risk), also set:

export ALPACA_PAPER="false"

3. Install the Alpaca SDK

pip install alpaca-trade-api

4. Paper Trading Commands

# See what trades would be made (no execution)
python paper_trade.py --dry-run

# Check current portfolio status
python paper_trade.py --status

# Execute rebalance (only runs if 21+ trading days since last)
python paper_trade.py

# Force rebalance regardless of schedule
python paper_trade.py --force

5. Automate with Cron

Add this to your crontab to auto-rebalance at 3:55 PM ET on weekdays:

crontab -e

55 15 * * 1-5 cd /path/to/trade && python paper_trade.py >> logs/trade.log 2>&1

The script tracks state in logs/paper_trade_state.json and only rebalances when the configured interval (21 trading days) has passed. Use --force to override.

6. Monitoring

• Trade logs: logs/paper_trade_YYYYMMDD.log — every order with timestamps and fill prices
• State file: logs/paper_trade_state.json — last rebalance date and full trade history
• Quick check: python paper_trade.py --status

---

Configuration

All parameters live in config.yaml:

universe:
  symbols: [AAPL, MSFT, GOOGL, ...]   # Stocks to trade
  benchmark: SPY                        # Benchmark index

portfolio:
  max_positions: 20          # Max holdings
  max_position_weight: 0.10  # 10% cap per stock
  min_position_weight: 0.02  # 2% floor per stock
  target_volatility: 0.15    # 15% annual vol target
  rebalance_frequency_days: 21
  transaction_cost_bps: 10

risk:
  max_drawdown_limit: 0.20   # Halt at 20% drawdown
  max_sector_weight: 0.30    # 30% sector cap
  stop_loss_pct: 0.08        # 8% per-stock stop loss

Customizing the Universe

Edit the symbols list in config.yaml. The system works with any US equities available on Yahoo Finance.

Adjusting Risk

• More conservative: Lower target_volatility (e.g., 0.10), lower max_position_weight, increase min_position_weight
• More aggressive: Raise target_volatility (e.g., 0.20), allow larger positions

---

Project Structure

trade/
├── config.yaml                 # Strategy configuration
├── run.py                      # CLI: backtest & signals
├── paper_trade.py              # Alpaca paper/live trading
├── quant_trading_colab.ipynb   # Interactive Colab notebook
├── requirements.txt
├── quant/
│   ├── strategy.py             # Main orchestrator
│   ├── data/
│   │   └── market_data.py      # Yahoo Finance data fetching
│   ├── signals/
│   │   └── factors.py          # Alpha factor calculations
│   ├── portfolio/
│   │   └── optimizer.py        # Mean-variance optimization
│   ├── backtest/
│   │   ├── engine.py           # Backtesting engine
│   │   └── report.py           # Analytics & reporting
│   ├── execution/
│   │   ├── broker.py           # Broker interface + paper broker
│   │   └── alpaca_broker.py    # Alpaca API integration
│   └── utils/
│       └── config.py           # Config loader
└── tests/                      # Unit tests

Disclaimer

This software is for educational and research purposes. Past performance does not guarantee future results. Use at your own risk. Always start with paper trading before committing real capital.