Country Manager & Data Science student at FGV EESP. Background in commercial leadership across Latin America — now building quantitative skills to combine with business experience.
Currently studying: Data Science (FGV EESP) — Econometrics · Time Series · Machine Learning · Python
| Project | Description | Stack |
|---|---|---|
| Returns to schooling in Brazil | Mincer equation · panel data 27 states · 2012–2025 · R²=0.94 | Python · statsmodels · linearmodels |
| BHP/VALE pairs trading | Cointegration · VECM · AR(2) spread · trading thresholds | R · vars · urca |
| Monte Carlo options pricing — PETR4 | GBM simulation · Longstaff-Schwartz · options valuation | Python · numpy · pandas |
| Markowitz efficient frontier — IBOV | SLSQP optimization · IPCA deflation · Monte Carlo · Sharpe 0.89 vs IBOV 0.17 | Python · scipy · python-bcb |
| Soy put hedge — Black-Scholes & backtesting | European put pricing · train/test split · floor validation · scenario analysis | Python · scipy · yfinance |
| Student stress classifier | Random Forest · SHAP explainability · 843 obs. | Python · scikit-learn · SHAP |
| IPCA forecast — aggregated vs. disaggregated | ARIMA · group-level forecasting · comparison with BCB Focus | R · forecast |
| Refund analytics dashboard | MTD comparison · KPI cards · deployed | Python · Streamlit |
| Bike sharing demand forecast | EDA · feature engineering · RF vs. linear baseline · RMSLE −56% | Python · scikit-learn |
- fgv-econometrics — OLS · Probit/Logit · IV · Panel data
- fgv-applied-time-series — VAR · VECM · ARCH/GARCH
- fgv-machine-learning — Ridge/Lasso · Logistic · ARIMA · Imputation
- fgv-python-data-analysis — Pandas · sklearn · visualization
- fgv-applied-statistics — Regression · hypothesis testing · GRETL
Python · R · SQL · pandas · scikit-learn · statsmodels · linearmodels · Streamlit · GRETL
📍 São Paulo, Brazil · LinkedIn