This project analyzes the impact of major geopolitical events on S&P 500 returns by calculating the Compound Annual Growth Rate (CAGR) at different time horizons after each event.
geo-risk-snp/
├── data/
│ ├── Geopolitical Risk v S&P500 returns - s&p500 daily returns 1950-2020.csv # Raw price data
│ └── cagr_results.csv # Calculated CAGR results
├── docs/
│ └── events.txt # List of geopolitical events
├── results/
│ ├── cagr_by_event.png # Bar chart visualization
│ ├── cagr_heatmap.png # Heatmap visualization
│ └── cagr_time_series.png # Time series visualization
├── src/
│ ├── preprocess.py # Script to preprocess the price data
│ ├── cagr_calculator.py # Script to calculate CAGR for each event
│ └── visualize_results.py # Script to visualize the results
├── tasks.md # Project tasks and pseudocode
└── README.md # This file
For each geopolitical event:
- The entry date is determined as 2 trading days after the event.
- The entry price is the S&P 500 Adjusted Close price on the entry date.
- CAGR is calculated for 1-year, 3-year, and 5-year horizons using the formula:
CAGR = (exit_price / entry_price) ^ (1 / years) - 1
The analysis has calculated CAGR for the following geopolitical events:
- Korean War Begins (1950)
- Suez Canal Crisis (1956)
- Cuban Missile Crisis (1962)
- JFK Assassinated (1963)
- Arab Oil Embargo (1973)
- President Nixon Resigns (1974)
- Iranian Hostage Crisis (1979)
- U.S.S.R. Invades Afghanistan (1979)
- U.S. Invades Panama (1989)
The results are visualized in three different formats:
- Bar chart comparing CAGR across events
- Heatmap showing CAGR by event and time horizon
- Time series showing how CAGR varies over time
-
To preprocess the data:
python src/preprocess.py -
To calculate CAGR:
python src/cagr_calculator.py -
To generate visualizations:
python src/visualize_results.py
- pandas
- numpy
- matplotlib
- seaborn