MissForest - non-parametric missing value imputation for mixed-type data

By Daniel J. Stekhoven, Peter Buhlmann. Bioinform., 2011

5,545 citations recorded by the source corpus.

MOTIVATION 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 non-parametric method which can cope with different types of variables simultaneously. RESULTS 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 datasets 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 datasets including different types of variables. In our comparative study, missForest outperforms other methods of imputation especially in data settings where complex interactions and non-linear 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. AVAILABILITY The package missForest is freely available from http://stat.ethz.ch/CRAN/. CONTACT stekhoven@stat.math.ethz.ch; buhlmann@stat.math.ethz.ch

Subfields: Molecular Biology, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/8bbab0d33bd065933d675ba1a8591a55353a70fe

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Molecular Biology

In Statistics and Probability

Citations in the corpus

Random forest predictive modeling of mineral prospectivity with small number of prospects and data with missing values in Abra (Philippines) (2015, methodology): “Readers are referred to Stekhoven and Bühlmann (2012) for details of the missForest imputation algorithm.”
Random forest missing data algorithms (2017, methodology): “A third approach is missForest, a method recently introduced in [21].”
Random Forests for Genomic Data Analysis (2012, methodology): “RecentlyStekhovenandBuhlmannintroducedanothermethod of imputation by predicting missing values using RF trained on non-missing data [60].”
Imputation with the R Package VIM (2016, methodology): “…and Giannerini 2016), as well as multiple imputation methods, e.g., in the R packages mice (van Buuren and Groothuis-Oudshoorn 2011), AmeliaII (Honaker, King, and Blackwell 2011), mix (Schafer 2015), missMDA (Josse and Husson 2016), Hmisc (Harrell 2015), missForest (Stekhoven and Bühlmann 2012).”
Assembly mechanisms determining high species turnover in aquatic communities over regional and continental scales (2016, methodology): “The R component consisted of a matrix with based on random forest to impute missing values (miss Forest; Stekhoven and Bühlmann 2012).”
Cardiovascular disease risk prediction using automated machine learning: A prospective study of 423,604 UK Biobank participants (2019, methodology): “We used the MissForest non-parametric data imputation algorithm [25] to recover the missing values.”
Understanding Random Forests: From Theory to Practice (2014, methodology): “As such, 4.5 consistency 81 the MissForest algorithm [Stekhoven and Bühlmann, 2012] consists in iteratively building a random forest on the observed parts of the data in order to predict the missing values for a given variable.”
GAIN: Missing Data Imputation using Generative Adversarial Nets (2018, neutral): “…for GAIN and 5 other state-of-the-art imputation methods: MICE (Buuren & Oudshoorn, 2000; Buuren & Groothuis-Oudshoorn, 2011), MissForest (Stekhoven & B¨uhlmann, 2011), Matrix completion (Matrix) (Mazumder et al., 2010a), Auto-encoder (Gondara & Wang, 2017) and Expectation-maximization…”
Comparison of Random Forest and Parametric Imputation Models for Imputing Missing Data Using MICE: A CALIBER Study (2014, methodology): “In addition to MICE, we also evaluated missForest (14), which uses random forest in an iterative way to complete a data set with missing values, where imputed values are equal to the random forest predictions rather than being randomly sampled from a conditional distribution.”
missMDA: A Package for Handling Missing Values in Multivariate Data Analysis (2016, methodology): “One of the most recent propositions for mixed data is that of Stekhoven and Bühlmann (2012) who suggest an approach based on random forests, which seems to outperform the other methods in terms of imputation quality.”