Final-year Econometrics & Operations Research master's student with a robust foundation in mathematical modelling, data science, and statistical programming. Experienced in advanced statistical analysis, with a proven ability to translate complex data into actionable business insights.
I'm a final-year Econometrics & Operations Research master's student at the Prague University of Economics and Business, with a minor in Data Engineering. My foundation blends rigorous mathematical modelling with practical, production-level data work.
At EGAP (Czech Export Bank's insurance arm) I worked inside the Risk Department — maintaining Solvency II capital models, calculating Underwriting SCR, and preparing regulatory documentation for the Czech National Bank.
At Trayan Company I built automated reporting pipelines, cleaned messy customer data, and created interactive dashboards that replaced hours of manual work every month.
(Econometrics & Operations Research)
End-to-end BI project on ~100k real Brazilian e-commerce orders. Integrated 9 relational tables into an analytical master table in Python, built a 4-page interactive Power BI dashboard, and mined GUHA association rules to find what drives bad customer reviews — delay over 30 days generates 63% bad reviews (lift 4.9×).
Analysed 10,300 users and 105,000 watch events across 6 relational tables through 12 business questions. Combined a Python ↔ Excel workflow — pivot-table contingency tables feeding chi-squared tests — with GUHA data mining and machine-learning models, predicting binge-watchers with 95.1% accuracy.
Analysed a historical loan portfolio of 10,000 clients across three distribution channels. Identified default drivers, evaluated channel profitability, and built a logistic regression classifier to predict loan defaults.
Built classification models to predict diabetes diagnosis from demographics, lifestyle habits, and clinical biomarkers across 100,000 patients. Compared Logistic Regression, Random Forest, and advanced pipelines with KNN imputation and cross-validation.
Compared GARCH and EGARCH models for Value at Risk estimation during market crisis periods. Applied the Dynamic Quantile Test to backtest VaR accuracy and assess whether volatility models adequately capture tail risk in extreme conditions.
Investigated the spatial impacts of population aging on regional employment across German and Austrian NUTS regions. Used spatial econometric models (spatial lag, spatial error) with Eurostat data to capture cross-regional dependency effects.
Whether you have a role, a dataset that needs untangling, or just want to talk data — reach out.