- Master 1 student in Economics, Data & Decision Science at SE2, Université Jean Monnet (Saint-Étienne)
- I work on data cleaning, statistical modelling, forecasting and dashboards
- Main interests: financial data, time series, econometrics, machine learning and decision theory
- Currently learning: experimental design in economics, Bayesian statistics, machine learning for finance
- Looking for an internship in data science or quantitative analysis
| Domain | Tools |
|---|---|
| Analysis & econometrics | Python (pandas, NumPy, statsmodels, SciPy, arch), R (tidyverse, forecast, tseries) |
| Networks & spatial | NetworkX, GeoPandas, rasterio, Leaflet |
| Machine learning & AI | scikit-learn, XGBoost, Transformers, LLM APIs (Claude, Qwen) |
| Visualisation & dashboards | Matplotlib, Seaborn, Plotly, Power BI, R Shiny / R Markdown |
| Data | SQL, Excel, Jupyter |
Quantitative finance & risk
| Project | Description | Stack |
|---|---|---|
| portfolio-optimization-risk | Backtest 2005 to 2018 of 4 portfolios and 5 VaR models. Minimum variance has 28% less volatility than 1/N | Python, SciPy, Ledoit-Wolf |
| garch-volatility-forecasting | GARCH, GJR and EGARCH on the S&P 500. GJR-GARCH-t gives the best out-of-sample forecasts | Python, arch |
| graph-systemic-risk-networkx | Correlation network of 10 global banks and shock simulation | Python, NetworkX |
African economies & spatial economics
| Project | Description | Stack |
|---|---|---|
| brvm-uemoa-macro-finance | Inflation 11 points lower in CFA countries than in their neighbours; efficiency tests of the BRVM | Python, statsmodels |
| cameroon-highway-network | Least-cost highway network for Cameroon (Faber, 2014), with Dijkstra and Kruskal written by hand | Python, GeoPandas |
| time-series-econometrics-r | ADF and KPSS tests, ARIMA forecasts of the CAC 40, ARDL model of unemployment and inflation | R |
AI, NLP & decision science
| Project | Description | Stack |
|---|---|---|
| llm-matching | LLM reads motivation letters, Gale-Shapley assigns the seats; compared with a direct LLM assignment | Python, LLM APIs |
| nlp-projects | Tokenization benchmark, text clustering, sentiment of a WhatsApp forum with BERT | Python, scikit-learn, spaCy, Transformers |
| StudECO_AI | Agent that summarises economics papers and courses with a fixed template and checks the summary | Python, Qwen, Streamlit |
Data analysis & dashboards
| Project | Description | Stack |
|---|---|---|
| ecommerce-scraping-dashboard | Scrapes 5 e-commerce sites and compares prices, promotions and ratings | Python, BeautifulSoup, Streamlit |
| world-population-dashboard | Interactive dashboard of global demographic trends and migration flows | R, Flexdashboard, Leaflet |
| r-statistical-projects | PCA of French cities, fertility rate regression, Shiny evaluation app | R, Shiny |
| sales-data-cleaning-analysis | Data quality audit, outlier handling and customer behaviour analysis | Python, pandas |