Open to opportunities

Eduardo Araque

@eduardoaraque

Message

Data science intern at ARCH who authored a SARIMAX forecasting methodology and built its Python analysis tool.

Ecuador

Message

What I'm looking for

I’m seeking a data science, analytics, or statistics role, and I’m open to remote, freelance, or hybrid work in Quito.

At ARCH, Ecuador’s hydrocarbon regulatory agency, I authored a 73-page methodology for forecasting petroleum product and LPG volumes. It proposes a traceable, reproducible statistical workflow using SARIMA/SARIMAX models, intervention analysis, and anomaly detection.

I also built the methodology’s Python tool with pandas and statsmodels, including an analyst notebook, reusable modules, and automatic Word report generation. I wrote its theoretical framework and user manuals, and created an Excel template with data-quality controls.

For a university community-outreach project with NOVA S.A., I developed a calibrated XGBoost model to predict next-day rain from humidity and pressure. After auditing the data and correcting critical preprocessing errors, I evaluated the model against baselines and advised against using it as an alert system because it did not outperform a 30-day climatology baseline.

My portfolio includes complex survey sampling with official Ecuadorian data, a linear regression implementation from scratch in NumPy, and a reproducible cleaning and analysis pipeline for INEC death records. I’m completing a mathematics degree at Central University of Ecuador and work with statistics, Python, and R.

Experience

Work history, roles, and key accomplishments

Statistics Intern

ARCH

Mar 2026 - Sep 2026 (6 months)

Authored a new Volume Programming Methodology for petroleum products and LPG, designing a SARIMA/SARIMAX-based statistical workflow and building a Python tool with reusable modules and automatic report generation.

Python Pandas Statsmodels SARIMA Intervention Analysis Box Cox Transformation Augmented Dickey Fuller (ADF) Test KPSS Test AICC Residual Diagnostics Anomaly Detection Time Series Forecasting Word Report Generation

Technical Lead

Central University of Ecuador

Oct 2025 - Feb 2026 (4 months)

Led a university community-outreach project developing a calibrated XGBoost classifier for next-day rain prediction, auditing preprocessing to fix critical errors and evaluating with strict temporal validation.

XGBoost Platt Scaling Feature Engineering Temporal Validation TimeSeriesSplit Walk Forward Validation Python scikit learn Data Auditing Repeatability Testing Automation

Education

Degrees, certifications, and relevant coursework

Central University of Ecuador

B.Sc. in Mathematics, Mathematics

2021 -

Activities and societies: Authored lecture notes on inference with R, covering probability sampling, Horvitz–Thompson estimator, and survey analysis with the survey package.

Final-year mathematics student with expected graduation in 2027. Key courses include Probability, Mathematical Statistics, Statistical Inference, Linear Models, Optimization, and Numerical Analysis.