Roderick J. A. Little · The Encyclopedia of Adulthood and Aging · 2015
Missing data are a common problem in adulthood and aging research. This entry overviews the problem and possible solutions. It begins by offering a definition of missing data, with examples. It presents a widely applied taxonomy of mechanisms that create the missing data. It then considers common, but limited, approaches: complete‐cases, available cases, weighting analyses, and single imputation. More principled methods, namely multiple imputation, maximum likelihood, and inference from Bayesian posterior distributions, are then discussed.
Cited by 3
Bhagyashri A. Kelkar · International Journal of Decision Support System Technology · 2021
Many real world datasets may contain missing values for various reasons. These incomplete datasets can pose severe issues to the underlying machine learning algorithms and decision support systems. It may result in high computational cost, skewed output and invalid deductions. Various solutions exist to mitigate this issue; the most popular strategy is to estimate the missing values by applying inferential techniques such as linear regression, decision trees or Bayesian inference. In this paper, the missing data problem is discussed in detail with a comprehensive review of the approaches to tackle it. The paper concludes with a discussion on the effectiveness of three imputation methods namely, imputation based on Multiple Linear Regression (MLR), Predictive Mean Matching (PMM) and Classification And Regression Tree (CART) in the context of subspace clustering. The experimental results obtained on real benchmark datasets and high-dimensional synthetic datasets highlight that, MLR based imputation method is more efficient on high-dimensional incomplete datasets.
Cited by 8
Ton J. Cleophas, Aeilko H. Zwinderman · SpringerBriefs in statistics · 2012
Cited by 2
M. Nguyen · 2025
Cited by 0
Jing, Ling · eScholarship (California Digital Library) · 2012
Existence of missing values creates a big problem in real world data. Unless those values are missing completely at random, we cannot disregard them. This paper demonstrates some implementation methods to deal with missing values. It will show the theory and implementations of EM Algorithm, Regression Imputation, Stochastic Regression Imputation and Multiple imputations. This paper begins by introducing the theories of those methods and then applying them to two examples and finally diagnostics. I will use some examples to demonstrate these algorithms and hope this helps researchers with various backgrounds solve missing values problem in their data.
Cited by 0Open access
Tibor V. Varga · Research at the University of Copenhagen (University of Copenhagen) · 2019
Offers a convenient pipeline to test and compare various missing data imputation algorithms on simulated and real data. These include simpler methods, such as mean and median imputation and random replacement, but also include more sophisticated algorithms already implemented in popular R packages, such as 'mi', described by Su et al. (2011) <<a href="https://doi.org/10.18637%2Fjss.v045.i02" target="_top">doi:10.18637/jss.v045.i02</a>>; 'mice', described by van Buuren and Groothuis-Oudshoorn (2011) <<a href="https://doi.org/10.18637%2Fjss.v045.i03" target="_top">doi:10.18637/jss.v045.i03</a>>; 'missForest', described by Stekhoven and Buhlmann (2012) <<a href="https://doi.org/10.1093%2Fbioinformatics%2Fbtr597" target="_top">doi:10.1093/bioinformatics/btr597</a>>; 'missMDA', described by Josse and Husson (2016) <<a href="https://doi.org/10.18637%2Fjss.v070.i01" target="_top">doi:10.18637/jss.v070.i01</a>>; and 'pcaMethods', described by Stacklies et al. (2007) <<a href="https://doi.org/10.1093%2Fbioinformatics%2Fbtm069" target="_top">doi:10.1093/bioinformatics/btm069</a>>. The central assumption behind 'missCompare' is that structurally different datasets (e.g. larger datasets with a large number of correlated variables vs. smaller datasets with non correlated variables) will benefit differently from different missing data imputation algorithms. 'missCompare' takes measurements of your dataset and sets up a sandbox to try a curated list of standard and sophisticated missing data imputation algorithms and compares them assuming custom missingness patterns. 'missCompare' will also impute your real-life dataset for you after the selection of the best performing algorithm in the simulations. The package also provides various post-imputation diagnostics and visualizations to help you assess imputation performance.
Cited by 0
Christopher J. Swearingen · 2014
Cited by 1
The SAGE Encyclopedia of Research Design · 2022
Cited by 0
Camil Fuchs, Ron S. Kenett · Wiley StatsRef: Statistics Reference Online · 2014
Abstract Handling missing data is a core requirement of data analysis. The assessment of the process that caused missing data and the approach for handling missing data can have significant impact on the results of the analysis. In this section, we address the issue of missing data and imputation, covering a wide variety of applications and solutions proposed in the recent literature.
Cited by 3
Amir Momeni, Matthew R. Pincus, Jenny Libien · 2017
Cited by 6
Qinbao Song, Martin Shepperd · International Journal of Business Intelligence and Data Mining · 2007
Intelligent data analysis techniques are useful for better exploring real-world data sets. However, the real-world data sets always are accompanied by missing data that is one major factor affecting data quality. At the same time, good intelligent data exploration requires quality data. Fortunately, Missing Data Imputation Techniques (MDITs) can be used to improve data quality. However, no one method MDIT can be used in all conditions, each method has its own context. In this paper, we introduce the MDITs to the KDD and machine learning communities by presenting the basic idea and highlighting the advantages and limitations of each method.
Cited by 72
Davoud Torkamaneh (768035), Francois Belzile (5664199) · Figshare · 2015
<p>(a) The accuracy of imputed missing data (in %) is plotted against the proportion of missing data (in %) tolerated (MaxMD) at three levels of minimal minor allele frequency (MinMAF). (b) Accuracy of overall GBS dataset (in %) after imputation at different levels of MaxMD and MinMAF.</p>
Cited by 0Open access
Meijs, A.S.G.M. van der · Theses database, Tilburg University · 2018
Cited by 0
Camil Fuchs, Ron S. Kenett · Encyclopedia of Statistics in Quality and Reliability · 2007
Abstract Handling missing data is a core requirement of data analysis. The assessment of the process that caused missing data and the approach for handling missing data can have significant impact on the results of the analysis. In this section, we address the issue of missing data and imputation, covering a wide variety of applications and solutions proposed in the recent literature.
Cited by 3
Gianluca Camillieri · 2024
Cited by 1
Alanis, Joseph Omar · ScholarWorks @ UTRGV (The University of Texas Rio Grande Valley) · 2024
In the repeated measures longitudinal datasets where missing data is a relatively common issue, we explored different imputation methods, including machine learning (ML) methods in order to examine potential efficiencies for using traditional versus newer computational methods. In order to accomplish said comparison, we used a Monte Carlo simulation experiment of a population of size N=70000 to mimic a clinical trial, with different scenarios of missing at random (MAR) data in the response variable. To compare the behavior of each method resulting from the difference between population and sample dataset, a real dataset was used from the Boston College on "National Longitudinal Survey" to simulate MAR. Moreover, we used both datasets to examine the effects of different sample sizes when using Bayesian neural networks and k-Nearest Neighbors (k-NN) for imputations and compared that to the more traditional methods of last observed carried forward, multiple imputation, and linear regression. Additionally, the cost of computing power is evaluated at different sample sizes for each scenario on both datasets.
Cited by 0
Karen C. Barrett, Nancy L. Leech, George A. Morgan · 2026
Cited by 0
Agostino Di Ciaccio · IRIS Research product catalog (Sapienza University of Rome) · 2015
With big administrative data, often we have a large number of variables with different measurement levels and many missing data. The correct approach to \nhandle these situations depends on the type of data and the purpose of analysis. However, we can not simply delete the incomplete records, because it amounts to \na substantial loss of costly collected data. Single imputation or multiple imputation can be applied to obtain different aims, create an ‘imputed’ data matrix with the \nsame characteristics of the observed data or take account, in the estimation of a model, of the additional variability due to the imputation process. For big administrative \ndata, several approaches have been proposed in literature. In this paper we compare different approaches, considering both single and multiple imputation, and \nwe propose a new method, named Multitree. By some simulations, we show that Multitree is competitive with the best methods considered in literature.
Cited by 0
Ton J. Cleophas, Aeilko H. Zwinderman · 2015
Cited by 0
Peter Cummings · JAMA Pediatrics · 2013
Missing data can result in biased estimates of the association between an exposure X and an outcome Y. Even in the absence of bias, missing data can hurt precision, resulting in wider confidence intervals. Analysts should examine the missing data pattern and try to determine the causes of the missingness. Modern software has simplified multiple imputation of missing data and the analysis of multiply imputed data to the point where this method should be part of any analyst's toolkit. Multiple imputation will often, but not always, reduce bias and increase precision compared with complete-case analysis. Some exceptions to this rule are noted in this review. When describing study results, authors should disclose the amount of missing data and other details. Investigators should consider how to minimize missing data when planning a study.
Cited by 180