A complete group data-analysis project (Group 5, Boolean Data Analytics) investigating the rise in global temperatures — the target of the Paris Agreement — and how it correlates with CO₂ emissions, population and GDP across countries.
Designed to stay current: it includes a script that periodically finds new relevant datasets to fold into the analysis.
| Stage | File | What it does |
|---|---|---|
| Correlation analysis | CorrelationTest_Group5.py |
Pearson correlation between annual mean temperature and CO₂ (MtCO₂e), population and GDP (1980–2019) |
| Significance testing | permutation_test_temperature.ipynb |
Permutation test validating the temperature trend isn't down to chance |
| Forecasting | RNNTest_Group5.py |
Recurrent neural network prototype for temperature time-series forecasting |
| Main analysis | main.ipynb |
The core exploratory notebook tying the analysis together |
| Dashboard | Global_Warming_Dashboard.twb |
Interactive Tableau visualization |
| Data & report | Global Warming Dateset_Group5.xlsx · Final_GROUP5.pdf |
The dataset we built + the presentation slides |
- Strong positive correlation between global mean temperature and cumulative CO₂ emissions (Pearson, validated via permutation test)
- Analysis window 1980–2019, cross-referenced against population and GDP per country
Python · pandas · SciPy (Pearson, permutation tests) · scikit-learn · RNN · Tableau · Excel
- Statistical rigor — not just correlation, but a permutation test to check significance
- Built for updates — a routine to pull fresh datasets so the analysis reflects current data
- Full storytelling — from raw data to a Tableau dashboard and a presented deck
Group 5 — Boolean Data Analytics. Repo maintained by Matteo Balducci — Data Analyst.