MissForest - nonparametric missing value imputation for mixed-type data

Abstract

Modern data acquisition based on high-throughput technology is often facing the problem of missing data. Algorithms commonly used in the analysis of such large-scale data often depend on a complete set. Missing value imputation offers a solution to this problem. However, the majority of available imputation methods are restricted to one type of variable only: continuous or categorical. For mixed-type data the different types are usually handled separately. Therefore, these methods ignore possible relations between variable types. We propose a nonparametric method which can cope with different types of variables simultaneously. We compare several state of the art methods for the imputation of missing values. We propose and evaluate an iterative imputation method (missForest) based on a random forest. By averaging over many unpruned classification or regression trees random forest intrinsically constitutes a multiple imputation scheme. Using the built-in out-of-bag error estimates of random forest we are able to estimate the imputation error without the need of a test set. Evaluation is performed on multiple data sets coming from a diverse selection of biological fields with artificially introduced missing values ranging from 10% to 30%. We show that missForest can successfully handle missing values, particularly in data sets including different types of variables. In our comparative study missForest outperforms other methods of imputation especially in data settings where complex interactions and nonlinear relations are suspected. The out-of-bag imputation error estimates of missForest prove to be adequate in all settings. Additionally, missForest exhibits attractive computational efficiency and can cope with high-dimensional data.

Submitted to Oxford Journal's Bioinformatics on 3rd of May 2011

References in corpus (1)

• Pattern Alternating Maximization Algorithm for Missing Data in Large P, Small N Problems

Cited by in corpus (131)

• A Methodology to Derive Global Maps of Leaf Traits Using Remote Sensing and Climate Data
• Deep learning cardiac motion analysis for human survival prediction
• Missing Data Imputation with Adversarially-trained Graph Convolutional Networks
• The effect of short-term exposure to the natural environment on depressive mood: A systematic review and meta-analysis
• Machine learning algorithms to infer trait-matching and predict species interactions in ecological networks
• Handling missing values in healthcare data: A systematic review of deep learning-based imputation techniques
• Classification of datasets with imputed missing values: does imputation quality matter?
• Graph Convolutional Networks for Graphs Containing Missing Features
• Construction of a Calibrated Probabilistic Classification Catalog: Application to 50k Variable Sources in the All-Sky Automated Survey
• Handling Missing Data with Graph Representation Learning
• Stable Graphical Model Estimation with Random Forests for Discrete, Continuous, and Mixed Variables
• Establishing strong imputation performance of a denoising autoencoder in a wide range of missing data problems
• A review of Generative Adversarial Networks for Electronic Health Records: applications, evaluation measures and data sources
• MedGCN: Medication recommendation and lab test imputation via graph convolutional networks
• Variational Autoencoder with Arbitrary Conditioning
• On the consistency of supervised learning with missing values
• Imputation of missing sub-hourly precipitation data in a large sensor network: a machine learning approach
• The shapes of an epidemic: using Functional Data Analysis to characterize COVID-19 in Italy
• Automated classification of Hipparcos unsolved variables
• Self-Attention Between Datapoints: Going Beyond Individual Input-Output Pairs in Deep Learning
• Data Generators for Learning Systems Based on RBF Networks
• AutoPrognosis 2.0: Democratizing Diagnostic and Prognostic Modeling in Healthcare with Automated Machine Learning
• Automatic Classification of Variable Stars in Catalogs with missing data
• STING: Self-attention based Time-series Imputation Networks using GAN
• A Deep Learning Approach for Overall Survival Prediction in Lung Cancer with Missing Values
• Characteristics of ChatGPT users from Germany: implications for the digital divide from web tracking data
• cleanTS: Automated (AutoML) Tool to Clean Univariate Time Series at Microscales
• Interpretable machine learning for high-dimensional trajectories of aging health
• Proper motion survey and kinematic analysis of the Rho Ophiuchi embedded cluster
• A data-driven prospective study of incident dementia among older adults in the United States
• not-MIWAE: Deep Generative Modelling with Missing not at Random Data
• A Robust and Flexible EM Algorithm for Mixtures of Elliptical Distributions with Missing Data
• Preparing for advanced LIGO: A Star-Galaxy Separation Catalog for the Palomar Transient Factory
• DC-Check: A Data-Centric AI checklist to guide the development of reliable machine learning systems
• Wind energy forecasting with missing values within a fully conditional specification framework
• Hierarchical inference for genome-wide association studies: a view on methodology with software
• proFIA: A data preprocessing workflow for Flow Injection Analysis coupled to High-Resolution Mass Spectrometry
• Comparison of Missing Data Imputation Methods using the Framingham Heart study dataset
• Missing Value Imputation Based on Deep Generative Models
• Imputation of missing values in multi-view data
• Deep Generative Imputation Model for Missing Not At Random Data
• Instructions and Guide for Diagnostic Questions: The NeurIPS 2020 Education Challenge
• A principal components method to impute missing values for mixed data
• On the Relation between Prediction and Imputation Accuracy under Missing Covariates
• Modern Multiple Imputation with Functional Data
• The Missing Indicator Method: From Low to High Dimensions
• Identifiable Generative Models for Missing Not at Random Data Imputation
• Are deep learning models superior for missing data imputation in large surveys? Evidence from an empirical comparison
• A Comparative Study of Imputation Methods for Multivariate Ordinal Data
• Inward and Outward Network Influence Analysis
• Imputations for High Missing Rate Data in Covariates via Semi-supervised Learning Approach
• Accounting for multiple imputation-induced variability for differential analysis in mass spectrometry-based label-free quantitative proteomics
• Assessing Fairness in the Presence of Missing Data
• Private Post-GAN Boosting
• TEE4EHR: Transformer Event Encoder for Better Representation Learning in Electronic Health Records
• Flow Models for Arbitrary Conditional Likelihoods
• A study of distributionally robust mixed-integer programming with Wasserstein metric: on the value of incomplete data
• Adaptive Optimization for Prediction with Missing Data
• Comprehensive Overview of Bottom-up Proteomics using Mass Spectrometry
• Entropy Balancing for Causal Generalization with Target Sample Summary Information
• When to Impute? Imputation before and during cross-validation
• Sample Selection Bias in Machine Learning for Healthcare
• Gaia Focused Product Release: A catalogue of sources around quasars to search for strongly lensed quasars
• AutoCure: Automated Tabular Data Curation Technique for ML Pipelines
• Multiple Imputation Through XGBoost
• Random Forest Missing Data Algorithms
• Student Log-Data from a Randomized Evaluation of Educational Technology: A Causal Case Study
• Mixture-based Multiple Imputation Model for Clinical Data with a Temporal Dimension
• OCL: Ordinal Contrastive Learning for Imputating Features with Progressive Labels
• Logistic Regression with Missing Covariates -- Parameter Estimation, Model Selection and Prediction within a Joint-Modeling Framework
• Blood lactate concentration prediction in critical care patients: handling missing values
• Model-assisted estimation through random forests in finite population sampling
• Development and Validation of a Deep Learning Model for Prediction of Severe Outcomes in Suspected COVID-19 Infection
• Clinical utility of foundation models in musculoskeletal MRI for biomarker fidelity and predictive outcomes
• Semi-supervised Learning with Missing Values Imputation
• Who wins the Miss Contest for Imputation Methods? Our Vote for Miss BooPF
• Prioritizing Variables for Observational Study Design using the Joint Variable Importance Plot
• Robust Sample Weighting to Facilitate Individualized Treatment Rule Learning for a Target Population
• Quality control, data cleaning, imputation
• Goodness (of fit) of Imputation Accuracy: The GoodImpact Analysis
• Characterizing User Susceptibility to COVID-19 Misinformation on Twitter
• Hierarchical Multi-Label Classification with Missing Information for Benthic Habitat Imagery
• Multi-View Variational Autoencoder for Missing Value Imputation in Untargeted Metabolomics
• Generative Adversarial Classification Network with Application to Network Traffic Classification
• Variable selection with missing data in both covariates and outcomes: Imputation and machine learning
• Variational Selective Autoencoder: Learning from Partially-Observed Heterogeneous Data
• Imputation and low-rank estimation with Missing Not At Random data
• Random Forests of Interaction Trees for Estimating Individualized Treatment Effects in Randomized Trials
• The Role of Mastery Learning in Intelligent Tutoring Systems: Principal Stratification on a Latent Variable
• Multiple Imputation with Neural Network Gaussian Process for High-dimensional Incomplete Data
• PC-GAIN: Pseudo-label Conditional Generative Adversarial Imputation Networks for Incomplete Data
• Multiple imputation using chained random forests: a preliminary study based on the empirical distribution of out-of-bag prediction errors
• Concept Drift Detection: Dealing with MissingValues via Fuzzy Distance Estimations
• Filling in the Blanks: Applying Data Imputation in incomplete Water Metering Data
• RIFLE: Imputation and Robust Inference from Low Order Marginals
• Gradient Importance Learning for Incomplete Observations
• Clustering and Variable Selection in the Presence of Mixed Variable Types and Missing Data
• Nonparametric imputation method for nonresponse in surveys
• DynImp: Dynamic Imputation for Wearable Sensing Data Through Sensory and Temporal Relatedness
• Discordance Minimization-based Imputation Algorithms for Missing Values in Rating Data
• Joints in Random Forests
• Correlating Visual Characteristics and Cryogenic Performance of Superconducting Detectors
• DPER: Efficient Parameter Estimation for Randomly Missing Data
• Influence of parallel computing strategies of iterative imputation of missing data: a case study on missForest
• Optimal Clustering with Missing Values
• A Projection Approach to Local Regression with Variable-Dimension Covariates
• On the Consistency of a Random Forest Algorithm in the Presence of Missing Entries
• A Data Imputation Model based on an Ensemble Scheme
• Projected Latent Markov Chain Monte Carlo: Conditional Sampling of Normalizing Flows
• Arbitrary Conditional Distributions with Energy
• Asymptotically Exact and Fast Gaussian Copula Models for Imputation of Mixed Data Types
• Beyond Accuracy: An Empirical Study of Uncertainty Estimation in Imputation
• Data-driven Lake Water Quality Forecasting for Time Series with Missing Data using Machine Learning
• What shapes climate change perceptions in Africa? A random forest approach
• Deep Multi-path Network Integrating Incomplete Biomarker and Chest CT Data for Evaluating Lung Cancer Risk
• Multiple Imputation with Massive Data: An Application to the Panel Study of Income Dynamics
• Discriminative, Generative and Self-Supervised Approaches for Target-Agnostic Learning
• Imputation techniques on missing values in breast cancer treatment and fertility data
• SDR-GAIN: A High Real-Time Occluded Pedestrian Pose Completion Method for Autonomous Driving
• Contrasting pre-vaccine COVID-19 waves in Italy through Functional Data Analysis
• Big Data Information and Nowcasting: Consumption and Investment from Bank Transactions in Turkey
• A Hamiltonian Monte Carlo Model for Imputation and Augmentation of Healthcare Data
• An utopic adventure in the modelling of conditional univariate and multivariate extremes
• Extremist ideology as a complex contagion: the spread of far-right radicalization in the United States between 2005-2017
• HMLasso: Lasso with High Missing Rate
• Modeling Extreme Events: Univariate and Multivariate Data-Driven Approaches
• MBExplainer: Multilevel bandit-based explanations for downstream models with augmented graph embeddings
• Cloned Identity Detection in Social-Sensor Clouds based on Incomplete Profiles
• Missingness Augmentation: A General Approach for Improving Generative Imputation Models
• Evaluation of individual attributes associated with shared HIV risk behaviors among two network-based studies of people who inject drugs
• Lung Cancer Risk Estimation with Incomplete Data: A Joint Missing Imputation Perspective