An interactive dashboard analysing how seven major economies responded to the post-pandemic inflation surge through monetary policy adjustments. Built as part of the CMT218 Data Visualisation module at Cardiff University (MSc Data Science).
The dashboard tracks policy rates, inflation, unemployment, and GDP growth across the United States, United Kingdom, Euro Area, Canada, Australia, Switzerland, and Japan from 2007 to 2030 (including IMF forecast data). It visualises the relationship between central bank rate decisions and macroeconomic outcomes during the most significant inflationary episode in decades.
- UK peaked at 11.1% inflation (Oct 2022) — the highest among all seven economies
- US hit 5.50% policy rate (2023) — the most aggressive tightening cycle
- Japan held at 0.10% throughout — a stark outlier driven by deflationary pressures
- Canada's 425 bps hike (Mar 2022–Jan 2023) was the fastest hiking cycle
Interactive Controls — Country toggle filters and dual-handle year range slider to explore specific economies and time periods.
Dual-Axis Line Chart — Inflation (solid) vs policy rate (dashed) trajectories with IMF forecast shading and key event annotations (GFC, COVID, inflation peak, rate cuts).
Inflation Intensity Heatmap — Cross-country inflation intensity over time with a diverging colour scale centred at the 2% target. Grouped Bar Chart — Pre-hike (2021), peak (2023), and current (2024) policy rate comparison.
GDP Growth Line Chart — Real data with IMF forecast horizon (dashed). Unemployment Lollipop Plot — Before-and-after comparison across three snapshots (pre-COVID, post-hike, latest).
Animated Scatter Plot — Dynamic bubble chart with Play/Pause controls. Each bubble represents one country in one year, with size mapped to unemployment rate and a 45° reference line marking zero real rates.
- Python 3.9+
- pip
git clone https://github.com/pedramebd/monetary-policy-dashboard.git
cd monetary-policy-dashboard
pip install -r requirements.txtpython app.pyThen open http://localhost:8050 in your browser.
To view the dashboard without any setup, open dashboard.html directly in any modern browser.
monetary-policy-dashboard/
├── app.py # Dash application (entry point)
├── Dataviz_Project.ipynb # Data cleaning, merging, and chart development
├── dashboard.html # Self-contained exported HTML dashboard
├── requirements.txt # Python dependencies
├── data/
│ ├── raw/ # Original source files (unmodified)
│ │ ├── policy_rates.xlsx
│ │ ├── consumer_prices.xlsx
│ │ ├── unemployment-rate.csv
│ │ └── gdp_growth.xls
│ └── processed/
│ └── dashboard_data.csv # Merged dataset (1,617 rows × 7 columns)
├── assets/ # Dashboard screenshots
├── .gitignore
├── LICENSE
└── README.md
| Dataset | Source | Format |
|---|---|---|
| Policy interest rates | Bank for International Settlements (BIS) | .xlsx |
| Consumer price inflation | Bank for International Settlements (BIS) | .xlsx |
| Unemployment rate | Our World in Data | .csv |
| GDP growth (incl. forecasts) | IMF World Economic Outlook | .xls |
Coverage: 7 economies, 2007–2030 (forecast years from IMF WEO).
- Python 3.9+
- Dash 4 — Web application framework
- Plotly — Interactive charting library
- dash-bootstrap-components — Layout and theming (FLATLY theme)
- Pandas — Data wrangling and merging
- xlrd / openpyxl — Excel file parsing
The Jupyter notebook documents the full pipeline:
- Ingestion — Loading four heterogeneous source files (CSV, XLS, XLSX)
- Cleaning — Standardising country names, date formats, and column schemas across sources
- Merging — Joining on country–year keys into a single tidy dataset (1,617 rows)
- Visualisation — Building six chart types with Plotly, iterating on design choices informed by visualisation research (Cleveland & McGill, Tufte, Munzner, Ware, Cairo)
- Deployment — Wrapping charts in a Dash 4 application with interactive controls and exporting a standalone HTML version
This project is released under the MIT License.
Pedram Ebadollahyvahed — MSc Data Science, Cardiff University (2025–2026)