Market Risk & Stress Testing Dashboard

An interactive Streamlit dashboard to analyze the market risk of equity/ETF portfolios.
It computes key risk metrics (volatility, drawdowns, Value at Risk, Expected Shortfall), visualizes portfolio vs benchmark performance, and runs simple stress tests and crisis‑window analyses.

Features

• Portfolio construction

  • Load historical prices for user‑selected tickers using yfinance (equities, ETFs, or indices).
  • Convert prices to daily returns and combine assets into a weighted portfolio.
  • Compute summary stats: daily/annualized volatility and maximum drawdown.

• Risk metrics

  • Historical Value at Risk (VaR) at configurable confidence levels (e.g., 95%, 99%).
  • Parametric (variance–covariance) VaR assuming normal returns using SciPy.
  • Historical Expected Shortfall (CVaR) as the average loss beyond VaR.

• Time‑varying correlation

  • Rolling correlation between portfolio returns and a chosen benchmark index.
  • Helps visualize regime changes when correlation spikes during crises.

• Performance visualization

  • Portfolio vs benchmark equity curves (cumulative return since start date).
  • Simple, clean charts built directly in Streamlit.

• Stress testing & crisis analysis

  • Uniform shock scenario: “What if the market drops X% in a day?” → approximate portfolio P&L for a user‑selected shock.
  • Historical crisis window (COVID‑like): slice returns into a fixed 2020 crash window and recompute drawdowns, VaR, and ES to show behavior under stress.

Tech Stack

• Python – core data handling and quantitative logic.
• Pandas / NumPy – time‑series processing, portfolio returns, volatility, and drawdown calculations.
• SciPy – normal distribution quantiles for parametric VaR.
• yfinance – download historical OHLCV price data from Yahoo Finance.
• Streamlit – web UI for inputs, charts, and interactive analysis.

Project Structure

market-risk-dashboard/
├─ app/
│  └─ main.py          # Streamlit app (UI + wiring to risk engine)
├─ risk_engine/
│  ├─ __init__.py
│  ├─ data.py          # Data download + return calculations
│  ├─ portfolio.py     # Portfolio returns + summary stats (vol, drawdown)
│  ├─ rolling.py       # Rolling correlation helper
│  ├─ var.py           # Historical & parametric VaR, Expected Shortfall
│  ├─ scenario.py      # Simple uniform shock stress test
│  ├─ crisis.py        # Crisis-window (e.g., COVID 2020) metrics
│  └─ tests/           # Quick scripts to test each module
└─ README.md

Getting Started

1. Clone the repository

git clone https://github.com/<your-username>/market-risk-dashboard.git
cd market-risk-dashboard

2. Create a virtual environment and install dependencies

python -m venv .venv
# Windows
.\.venv\Scripts\Activate
# Linux / macOS
source .venv/bin/activate

pip install -r requirements.txt  # if you create one
# or install manually:
# pip install pandas numpy yfinance streamlit scipy plotly

3. Run the Streamlit app

streamlit run app/main.py

Open the local URL shown in the terminal (usually http://localhost:8501).

How to Use

1. Enter tickers (comma‑separated), e.g. AAPL, MSFT, SPY.
2. Choose a start and end date for historical data.
3. Provide portfolio weights in the text area, one per line, e.g.:

   AAPL:0.3
   MSFT:0.3
   SPY:0.4
   

4. Select a benchmark ticker from the dropdown and pick a VaR confidence level.
5. Click Run analysis to:
   • See portfolio vs benchmark equity curves.
   • View volatility, drawdown, VaR, and ES.
   • Inspect rolling correlation with the benchmark.
   • Run a one‑day shock scenario and view COVID‑window metrics.

Possible Extensions

• Add rolling beta per asset vs benchmark.
• Support FX pairs or other asset classes.
• Add Monte Carlo or GARCH‑based VaR/ES models.
• Deploy on Streamlit Community Cloud or another hosting platform.