SINGLE IMPUTATION METHOD TO HANDLE MISSING VALUES IN LARGE DATASET

Background:Missing data in clinical epidemiological researches violate the intention to treat principle,reduce statistical power and can induce bias if they are related to patient’s response to treatment. In multiple imputation (MI), covariates are included in the imputation equation to predict the values of missing data.Objectives:To find the best approach to estimate and impute the missing values in Kuwait Registry for Rheumatic Diseases (KRRD) patients data.Methods:A number of methods were implemented for dealing with missing data. These includedMultivariate imputation by chained equations(MICE),K-Nearest Neighbors(KNN),Bayesian Principal Component Analysis(BPCA),EM with Bootstrapping(Amelia II),Sequential Random Forest(MissForest) and mean imputation. Choosing the best imputation method wasjudged by the minimum scores ofRoot Mean Square Error(RMSE),Mean Absolute Error(MAE) andKolmogorov–Smirnov D test statistic(KS) between the imputed datapoints and the original datapoints that were subsequently sat to missing.Results:A total of 1,685 rheumatoid arthritis (RA) patients and 10,613 hospital visits were included in the registry. Among them, we found a number of variables that had missing values exceeding 5% of the total values. These included duration of RA (13.0%), smoking history (26.3%), rheumatoid factor (7.93%), anti-citrullinated peptide antibodies (20.5%), anti-nuclear antibodies (20.4%), sicca symptoms (19.2%), family history of a rheumatic disease (28.5%), steroid therapy (5.94%), ESR (5.16%), CRP (22.9%) and SDAI (38.0%), The results showed that among the methods used, MissForest gave the highest level of accuracy to estimate the missing values. It had the least imputation errors for both continuous and categorical variables at each frequency of missingness and it had the smallest prediction differences when the models used imputed laboratory values. In both data sets, MICE had the second least imputation errors and prediction differences, followed by KNN and mean imputation.Conclusion:MissForest is a highly accurate method of imputation for missing data in KRRD and outperforms other common imputation techniques in terms of imputation error and maintenance of predictive ability with imputed values in clinical predictive models. This approach can be used in registries to improve the accuracy of data, including the ones for rheumatoid arthritis patients.References:[1]Junninen, H.; Niska, H.; Tuppurainen, K.; Ruuskanen, J.; Kolehmainen, M. Methods for imputation ofmissing values in air quality data sets.Atmospheric Environment2004,38, 2895–2907.[2]Norazian, M.N.; Shukri, Y.A.; Azam, R.N.; Al Bakri, A.M.M. Estimation of missing values in air pollutiondata using single imputation techniques.ScienceAsia2008,34, 341–345.[3]Plaia, A.; Bondi, A. Single imputation method of missing values in environmental pollution data sets.Atmospheric Environment2006,40, 7316–7330.[4]Kabir, G.; Tesfamariam, S.; Hemsing, J.; Sadiq, R. Handling incomplete and missing data in water networkdatabase using imputation methods.Sustainable and Resilient Infrastructure2019, pp. 1–13.[5]Di Zio, M.; Guarnera, U.; Luzi, O. Imputation through finite Gaussian mixture models.ComputationalStatistics & Data Analysis2007,51, 5305–5316.Disclosure of Interests:None declared

Download Full-text

Handling Complex Missing Data Using Random Forest Approach for an Air Quality Monitoring Dataset: A Case Study of Kuwait Environmental Data (2012 to 2018)

International Journal of Environmental Research and Public Health ◽

10.3390/ijerph18031333 ◽

2021 ◽

Vol 18 (3) ◽

pp. 1333

Author(s):

Ahmad R. Alsaber ◽

Jiazhu Pan ◽

Adeeba Al-Hurban

Keyword(s):

Air Quality ◽

Missing Data ◽

Random Forest ◽

Missing Values ◽

Imputation Method ◽

Environmental Data ◽

Environmental Research ◽

Quality Data ◽

Data Set ◽

Air Quality Data

In environmental research, missing data are often a challenge for statistical modeling. This paper addressed some advanced techniques to deal with missing values in a data set measuring air quality using a multiple imputation (MI) approach. MCAR, MAR, and NMAR missing data techniques are applied to the data set. Five missing data levels are considered: 5%, 10%, 20%, 30%, and 40%. The imputation method used in this paper is an iterative imputation method, missForest, which is related to the random forest approach. Air quality data sets were gathered from five monitoring stations in Kuwait, aggregated to a daily basis. Logarithm transformation was carried out for all pollutant data, in order to normalize their distributions and to minimize skewness. We found high levels of missing values for NO2 (18.4%), CO (18.5%), PM10 (57.4%), SO2 (19.0%), and O3 (18.2%) data. Climatological data (i.e., air temperature, relative humidity, wind direction, and wind speed) were used as control variables for better estimation. The results show that the MAR technique had the lowest RMSE and MAE. We conclude that MI using the missForest approach has a high level of accuracy in estimating missing values. MissForest had the lowest imputation error (RMSE and MAE) among the other imputation methods and, thus, can be considered to be appropriate for analyzing air quality data.

Download Full-text

A data-driven missing value imputation approach for longitudinal datasets

Artificial Intelligence Review ◽

10.1007/s10462-021-09963-5 ◽

2021 ◽

Author(s):

Caio Ribeiro ◽

Alex A. Freitas

Keyword(s):

Missing Data ◽

Longitudinal Data ◽

Missing Values ◽

Error Rates ◽

Imputation Method ◽

Data Driven ◽

Missing Value ◽

Missing Value Imputation ◽

Human Ageing ◽

Imputation Approach

AbstractLongitudinal datasets of human ageing studies usually have a high volume of missing data, and one way to handle missing values in a dataset is to replace them with estimations. However, there are many methods to estimate missing values, and no single method is the best for all datasets. In this article, we propose a data-driven missing value imputation approach that performs a feature-wise selection of the best imputation method, using known information in the dataset to rank the five methods we selected, based on their estimation error rates. We evaluated the proposed approach in two sets of experiments: a classifier-independent scenario, where we compared the applicabilities and error rates of each imputation method; and a classifier-dependent scenario, where we compared the predictive accuracy of Random Forest classifiers generated with datasets prepared using each imputation method and a baseline approach of doing no imputation (letting the classification algorithm handle the missing values internally). Based on our results from both sets of experiments, we concluded that the proposed data-driven missing value imputation approach generally resulted in models with more accurate estimations for missing data and better performing classifiers, in longitudinal datasets of human ageing. We also observed that imputation methods devised specifically for longitudinal data had very accurate estimations. This reinforces the idea that using the temporal information intrinsic to longitudinal data is a worthwhile endeavour for machine learning applications, and that can be achieved through the proposed data-driven approach.

Download Full-text

HIOC: a hybrid imputation method to predict missing values in medical datasets

International Journal of Intelligent Computing and Cybernetics ◽

10.1108/ijicc-03-2021-0042 ◽

2021 ◽

Vol ahead-of-print (ahead-of-print) ◽

Author(s):

Pooja Rani ◽

Rajneesh Kumar ◽

Anurag Jain

Keyword(s):

Breast Cancer ◽

Heart Disease ◽

Missing Values ◽

Imputation Method ◽

Support Vector ◽

Correct Prediction ◽

Breast Cancer Dataset ◽

Cancer Dataset ◽

Content Type ◽

Imputation Methods

PurposeDecision support systems developed using machine learning classifiers have become a valuable tool in predicting various diseases. However, the performance of these systems is adversely affected by the missing values in medical datasets. Imputation methods are used to predict these missing values. In this paper, a new imputation method called hybrid imputation optimized by the classifier (HIOC) is proposed to predict missing values efficiently.Design/methodology/approachThe proposed HIOC is developed by using a classifier to combine multivariate imputation by chained equations (MICE), K nearest neighbor (KNN), mean and mode imputation methods in an optimum way. Performance of HIOC has been compared to MICE, KNN, and mean and mode methods. Four classifiers support vector machine (SVM), naive Bayes (NB), random forest (RF) and decision tree (DT) have been used to evaluate the performance of imputation methods.FindingsThe results show that HIOC performed efficiently even with a high rate of missing values. It had reduced root mean square error (RMSE) up to 17.32% in the heart disease dataset and 34.73% in the breast cancer dataset. Correct prediction of missing values improved the accuracy of the classifiers in predicting diseases. It increased classification accuracy up to 18.61% in the heart disease dataset and 6.20% in the breast cancer dataset.Originality/valueThe proposed HIOC is a new hybrid imputation method that can efficiently predict missing values in any medical dataset.

Download Full-text

Estimating Semi-Parametric Missing Values with Iterative Imputation

International Journal of Data Warehousing and Mining ◽

10.4018/jdwm.2010070101 ◽

2010 ◽

Vol 6 (3) ◽

pp. 1-10 ◽

Cited By ~ 3

Author(s):

Shichao Zhang

Keyword(s):

Prior Knowledge ◽

Efficient Method ◽

Missing Values ◽

Imputation Accuracy ◽

Imputation Method ◽

Imputation Methods ◽

Real Dataset ◽

Regression Imputation ◽

Target Values ◽

Nonparametric Imputation

In this paper, the author designs an efficient method for imputing iteratively missing target values with semi-parametric kernel regression imputation, known as the semi-parametric iterative imputation algorithm (SIIA). While there is little prior knowledge on the datasets, the proposed iterative imputation method, which impute each missing value several times until the algorithms converges in each model, utilize a substantially useful amount of information. Additionally, this information includes occurrences involving missing values as well as capturing the real dataset distribution easier than the parametric or nonparametric imputation techniques. Experimental results show that the author’s imputation methods outperform the existing methods in terms of imputation accuracy, in particular in the situation with high missing ratio.

Download Full-text

A Safe-Region Imputation Method for Handling Medical Data with Missing Values

Symmetry ◽

10.3390/sym12111792 ◽

2020 ◽

Vol 12 (11) ◽

pp. 1792

Author(s):

Shu-Fen Huang ◽

Ching-Hsue Cheng

Keyword(s):

Missing Values ◽

Medical Data ◽

Imputation Method ◽

Attribute Selection ◽

Multivariate Normal ◽

Imputation Methods ◽

Safe Region ◽

Imbalanced Classes ◽

Stroke Trial ◽

Better Than

Medical data usually have missing values; hence, imputation methods have become an important issue. In previous studies, many imputation methods based on variable data had a multivariate normal distribution, such as expectation-maximization and regression-based imputation. These assumptions may lead to deviations in the results, which sometimes create a bottleneck. In addition, directly deleting instances with missing values may have several problems, such as losing important data, producing invalid research samples, and leading to research deviations. Therefore, this study proposed a safe-region imputation method for handling medical data with missing values; we also built a medical prediction model and compared the removed missing values with imputation methods in terms of the generated rules, accuracy, and AUC. First, this study used the kNN imputation, multiple imputation, and the proposed imputation to impute the missing data and then applied four attribute selection methods to select the important attributes. Then, we used the decision tree (C4.5), random forest, REP tree, and LMT classifier to generate the rules, accuracy, and AUC for comparison. Because there were four datasets with imbalanced classes (asymmetric classes), the AUC was an important criterion. In the experiment, we collected four open medical datasets from UCI and one international stroke trial dataset. The results show that the proposed safe-region imputation is better than the listing imputation methods and after imputing offers better results than directly deleting instances with missing values in the number of rules, accuracy, and AUC. These results will provide a reference for medical stakeholders.

Download Full-text