Diogo Ribeiro

Data Science & Research Lead writing practical notes on statistics, machine learning, forecasting, and data engineering. The useful posts here should add analysis, examples, code, experiments, or field judgment rather than paraphrase search results.

This page collects papers, draft manuscripts, reports and reproducible research repositories that have public source material. Status labels are conservative: a repository is listed as a draft or reproducible report unless the public metadata clearly identifies a publication.

Manuscripts and Drafts

Work Status Repository Manuscript
Stochastic diffusion processes, Brownian paths and implicit volatility coefficients Draft manuscript sml_diffusions_paper PDF, LaTeX
Article TeX repository Draft manuscript source article_tex_repo PDF, LaTeX
Short-rate innovations, monetary regimes and stock-market anomalies Reproducible manuscript short-rate-anomaly-regimes PDF, LaTeX
Regime-conditioned dynamical EVT for predictive maintenance Draft manuscript dynamical-evt-pdm PDF, LaTeX
Behavioral sensing models and change-point workflows Draft paper note behavioral-sensing-research Paper markdown, bibliography
BMSSP-style shortest-path implementation notes Draft research note bmssp Paper markdown, bibliography

Reproducible Reports

Work Status Repository Source
Minimum wages, productivity and inflation in Portugal, 1974-present Reproducible empirical report portugal-minimum-wage-inflation LaTeX, bibliography
Configurational pathways to firm innovation across European macroregions Reproducible fsQCA study europe-fsqca-innovation LaTeX, bibliography
Portugal 2025 turnout and OECD trust data Public article/data package democracy_missing_voters repository README
Portuguese external economic linkages, 1960-1973 Reproducible research pipeline portugal-external-growth-1960-1973 repository README
Portugal’s public-debt interest burden Reproducible analysis portugal-public-debt-interest repository README

Research-Code Companions

These repositories support or reproduce research arguments but are primarily code companions rather than standalone manuscripts.

  • brat-boosting — Python implementation of Boulevard Regularized Additive regression Trees with statistical inference.
  • sssp_research_lab — Implementable shortest-path algorithms inspired by sorting-barrier SSSP research.
  • adaptive-importance-sampling — Safe cross-entropy-based importance sampling for rare-event simulations.
  • ann-variable-selection-refutation — Reproduction and stress test of ANN variable selection from t-statistics.
  • pca-cautionary-tale — Reproducible PCA reliability workflow using PCA, t-SNE, persistent homology and cosine-distance modelling.

Inclusion Rule

I list public repositories when they contain at least one of: manuscript source, PDF output, bibliography, notebooks, replication code, or a documented research report. Private repositories and unpublished work without public source are intentionally excluded.