mcar

PyGrinder: a Python toolkit for grinding data beans into the incomplete for real-world data simulation by introducing missing values with different missingness patterns, including MCAR (complete at random), MAR (at random), MNAR (not at random), sub sequence missing, and block missing

Time-series missingness simulation separating mechanisms (MCAR/MAR/MNAR) from patterns (pointwise/block/monotone/decay/markov). Weighted multi-driver MAR, exact rate control, automatic calibration, 2D/3D support. Built for reproducible imputation benchmarking.

A library for synthetic missing data generation.

How Different Types of Missingness affect a complete Dataset

Healthcare missing-data analysis comparing complete-case, KNN, Bayesian iterative, and multiple imputation under MCAR, MAR, and MNAR.

A package for creating missing and faulty data

Reproducible data science pipeline to evaluate whether physicochemical water parameters can predict potability, including missingness mechanism assessment, cleaning, SQL exploration, and non-linear modeling.

Este estudio investiga la efectividad de la imputación múltiple en el análisis factorial confirmatorio (AFC) con datos de liderazgo, donde se simularon valores perdidos (MCAR) en un 40% de la muestra.

A comparative study of missing data imputation methods across MCAR, MAR, and MNAR mechanisms using classical, probabilistic, and deep learning approaches.

FAI (Feature-Wise Adaptive Imputation) is a machine learning framework that automatically selects the best imputation method per feature based on statistical properties — optimizing for downstream predictive performance, not just imputation error.