sampling-bias

An interactive framework to visualize and analyze your AutoML process in real-time.

Official Code for "Non-Probability Sampling Network for Stochastic Human Trajectory Prediction (CVPR 2022)"

Faster, better, smarter ecological niche modeling and species distribution modeling

Portfolio-grade audit of a student mental health & academic pressure survey. Measures coverage and sample imbalance, runs validity checks, highlights measurement and selection bias risks, and converts messy open-text “stress causes” into a transparent taxonomy. Ships a Markdown report, figures, and a Streamlit dashboard.

Longform data analysis article arguing every “dataset” is actually three: Observed (captured rows), Missing (what should exist but doesn’t), and Excluded (what filters/joins/dropna removed). Includes dataset accounting, join-loss and missingness audits, segmentation checks, and practical templates to prevent biased KPIs and wrong conclusions.

R code used for the analyses of the paper: Spatial conservation prioritisation in data-poor countries: a quantitative sensitivity analysis using different taxa

Forward-time simulation framework for studies of assortative mating and participation / ascertainment bias

Pipelines to evaluate Breast Cancer Purity Score and to correct sampling bias

Efficient Multistream Classification using Direct DensIty Ratio Estimation

This project presents an exploratory and statistical analysis of amphibian occurrence records in Brazil using data from the Global Biodiversity Information Facility (GBIF).

A global, taxon-stratified, high-resolution sampling-effort dataset from GBIF for bias-aware ecological modelling — reproducible R-based workflow for generating global sampling-effort rasters from GBIF occurrence records across spatial, temporal, and taxonomic scales. For more details, see El-Gabbas (2026), Diversity and distributions (accepted).

🚀📐 Representación gráfica de distribución de muestreos aleatorios.

Analysis of the accuracy of U.S. presidential election polls (2008–2024) based on prediction error data. Compares ideal vs. practical polling methodology, presents three contrasting visualizations of the same dataset to illustrate framing effects, and identifies selection/measurement biases with recommendations to improve accuracy.

Simulate the impact of sampling bias in the derivation of effect sizes

Overcoming Selection Bias in Statistical Studies With Amortized Bayesian Inference

📊 Evaluate survey quality and bias through coverage, representativeness, and measurement risk audits for reliable insights and data validity checks.