Transparent statistical pipeline for identifying biomarkers associated with neurological disease progression. Spearman correlations, Mann-Whitney group comparisons, volcano plots, and FDR correction.

Statistical analysis of 15 global airline stocks during COVID-19's two waves (2020) — using Spearman correlation, Wilcoxon Signed-Rank Test, and trend analysis in Excel to examine how adjusted close prices and trading volumes shifted between the panic and recovery phases.

A consolidated, multi-domain data science repository showcasing advanced data wrangling, statistical modeling, text mining, and machine learning. This Master's-level portfolio spans seven distinct projects—tackling challenges in FinTech, biostatistics, aviation, and developer ecosystems—using Python, Pandas, NumPy, & Scikit-Learn.

Transversal project in R: Biostatistical análisis applied to genetic expression

Exploratory data analysis identifying the customer behavioral drivers of annual revenue for an e-commerce platform using Python. Performed correlation analysis across customer metrics to surface the strongest revenue predictors and inform data-driven business recommendations.

Statistical factor research platform with Spearman Rank IC, quantile analysis, long-short diagnostics, and reproducible validation.

Предиктивный анализ оттока клиентов

Jupyter Notebook applying statistical theory to housing price prediction, using techniques like the Kolmogorov–Smirnov test, Box-Cox transformation, Pearson/Spearman correlations, chi-square tests, and feature importance, with analysis of prediction accuracy across price ranges.

Machine Learning solutions for a fintech. Improving the success rate of term deposits

Does pediatric cancer research activity align with survivor harm? A Simpson’s Paradox case study using ~15,000 ClinicalTrials.gov records — shows how disease-category aggregation can manufacture a false correlation that vanishes under subtype-level analysis.

4. Korelacija / zavisnost Pearson, Spearman, Kendall, partial correlation, distance correlation, Hoeffding D, MIC/MINE, maximal correlation, mutual information, conditional MI, normalized MI, pointwise MI (PMI), HSIC, copula (Gaussian, Clayton, Gumbel, Frank, t-copula), rank correlation matrices

Reproducibility package for a study linking regional U.S. dairy nutrition (DMI, NDF, hemicellulose) and farm economics to greenhouse gas emission intensity across nine major U.S. dairy production regions (2020).

Data mining analysis of F1 qualifying performance as a predictor of race outcomes (2006–2023). Includes Spearman correlation, contingency analysis, OLR, and a novel Team Reliability vs Pace quadrant framework.

Airline Customer Satisfaction Predictive Modelling using Python, Jupyter Notebook.

Gut microbiome co-occurrence network (graph mining): Spearman correlations + Benjamini-Hochberg FDR on 675 mouse gut samples, Louvain communities, LFPP vs Western diet comparison. Gephi exports included

Correlation-based network analysis comparing CWL vs. BWL conditions with sign-flip permutation inference (paired design)

Tiny dependency-free calculator: compute the Pearson & Spearman rank Information Coefficient of a trading signal from predictions vs realized returns. By microalphas.com

Java implementation for correlation coefficients

Statistical analysis of birthweight and related predictors using Python

analyse student exam performance using nonparametric statistical methods