🔗 Live Demo: https://portfolio-analyzer-rl.streamlit.app

Portfolio Analyzer (CLI)

A Python-based command-line tool for analyzing investment portfolios using live market data.
The tool computes portfolio allocation, unrealized profit/loss, benchmark performance, and risk-adjusted returns.

This project was built to explore how financial advisors and portfolio managers evaluate performance relative to the market.

---

Features

• Ingests portfolio holdings from a CSV file
• Pulls live and historical market prices using Yahoo Finance
• Computes:
  • Market value per holding
  • Portfolio allocation percentages
  • Cost basis and unrealized profit/loss
• Compares portfolio performance against a market benchmark (e.g., S&P 500 via SPY or VOO)
• Calculates risk-adjusted performance using the Sharpe ratio
• Fully configurable via command-line flags

---

Tech Stack

• Python
• pandas
• yfinance
• argparse (CLI interface)

---

Input Format

Holdings are provided via a CSV file with the following columns:

ticker,shares,avg_cost AAPL,5,150 MSFT,2,310 VOO,1,410

Run the analyzer from the project root:

python src/analyze.py

Common CLI Options:

python src/analyze.py --sort pl
python src/analyze.py --top 3
python src/analyze.py --benchmark SPY --period 1y
python src/analyze.py --risk_free 0.02

Available Flags:

• --file : Path to holdings CSV

• --sort : Sort output by allocation, unrealized P/L, or ticker

• --top : Show only the top N holdings

• --benchmark : Benchmark ticker (default: VOO)

• --period : Lookback period (e.g., 6mo, 1y, 2y)

• --risk_free : Annual risk-free rate (default: 0.0)

Sample Output:

Portfolio return (1y): +11.77%

SPY return (1y): +17.38%

Relative performance: -5.61%

Portfolio Sharpe (1y): 0.55

SPY Sharpe (1y): 0.82

ASSUMPTIONS AND LIMITATIONS:

• Portfolio returns are calculated as a weighted average of individual holding returns using current allocation weights

• Sharpe ratio is computed using daily returns and annualized

• This tool does not account for transaction timing, dividends, or taxes

These assumptions are appropriate for quick performance analysis but are not a full backtesting system.

---

Dashboard (Streamlit) + CLI Flags + Caching

This project includes both:

1. a CLI tool (engineer-friendly, scriptable), and
2. a web dashboard (demo-friendly, visually polished).

The dashboard also supports a CLI flags input box, so you can paste flags like:

benchmark SPY --period 2y --risk_free 0.02 --sort pl --top 5

Streamlit Caching (Stability Feature)

To reduce repeated calls to Yahoo Finance (yfinance) and improve reliability (especially when deployed), the dashboard caches analysis results for 15 minutes using Streamlit caching: • prevents rate-limit issues • speeds up refreshes and UI changes • improves stability for public users

Run Locally

1. Setup environment:

    python3 -m venv .venv
    source .venv/bin/activate
   

2. Install dependencies:

   pip install -r requirements.txt
   

3. Run the CLI Recommended way (package mode):

   python -m src.analyze --file data/holdings.csv --benchmark SPY --period 1y --risk_free 0.02
   Examples:
   python -m src.analyze --sort pl
   python -m src.analyze --top 5
   python -m src.analyze --benchmark SPY --period 2y --risk_free 0.02
   

4. Run the Dashboard:

   streamlit run app.py
   

---

Motivation:

This project was created to better understand how wealth management platforms evaluate portfolio performance, benchmark comparisons, and risk-adjusted returns in real-world financial contexts.