Search

Data sets with missing values bring great challenges to k-means clustering (KMC). At present, most studies focus on KMC data with low missing ratio while few studies on KMC data with high missing ratio. The current imputation methods have the following problems when dealing with the KMC data: (1) the error between imputation value and original true value is large, which leads to the poor imputation precision; (2) the imputation results have a great influence on the clustering results, which reduce the accuracies of the clustering results. We propose a novel imputation method, to deal with the problems, called as trimmed scores regression (TSR), which obtains an imputation estimator from a regression equation with a trimmed score matrix, and a novel cluster with k-means method. Compared with other imputation methods in numerical analysis, the TSR method exhibits better performance. [ABSTRACT FROM AUTHOR]

Dealing with missing data problems for skewed data is a difficult task especially since most of the imputation and data augmentation methodologies assume multivariate normality. The performance of imputation and hence the accuracy of inference on parameters become questionable when the violation of the above assumption occurs. One approach to solve the normality violation is to apply normalizing transformation prior to the imputation phase. However, this approach may introduce new problems such as altering the dependence structure among random variables. We present a general purpose multiple imputation approach based on Copula transformation. The approach is used to effectively transform any continuous multivariate non-normal data to multivariate normal, thereby allowing the imputation using standard normality-based techniques. The method then allows to conveniently back transform the data into original space. Real data sets are used to illustrate the techniques. We then compare the performance of our Copula-based method with other traditional normality-based multiple imputation approaches through extensive simulated and real non-normal multivariate datasets. We demonstrate that this method significantly mitigates the problem and hence the practice of making the blind assumption of multivariate normality for non-normal multivariate data under the assumption that data are missing at different mechanisms. [ABSTRACT FROM AUTHOR]

Missing data is an obligatory phenomenon in most of the sample surveys and when it is not handled cautiously at beginning of the study, outcomes may result to substantial biases in the survey estimates. Retaining such evidences into cognizance, this paper proffers few possible improved imputation methods to figure out the problem of missing data at the starting of the study and proffers computation procedure of the population mean using simple random sampling. The characteristics of the proffered imputation methods are ascertained till first order approximation pursued by a computational study carried out using several existent and hypothetical data sets. The findings are turned out rather promising showing dominance over the existing imputation methods. Further, apposite advices have been forwarded to the survey professionals for practical applications. [ABSTRACT FROM AUTHOR]

Missing data is common in survival analysis. It is either removed or imputed using various methods. Expectation-maximization (EM) imputation is a popular method in Cox regression studies. This paper investigated the effect of different regression methods on Cox regression modeling within the framework of EM. A stratified Cox regression model was derived from a dataset of categorical and numerical variables. Missing data were imputed using the EM framework with five machine learning algorithms and then were compared to the full model. The results show that the recursive partition and regression tree (RPART) method performed better than others. However, all regression methods performed poorly in categorical covariate imputation. R code is available online. [ABSTRACT FROM AUTHOR]

This article is an attempt to develop an effective stratified successive sampling estimation procedure which may cope with the problems of simultaneous presence of scrambled response and random non-response in realistic surveys. A generalized class of estimators has been suggested using imputation tool for reducing negative impacts of non-responses in sample surveys. Several estimators are presented as members of the suggested family. We have studied the properties of the suggested strategy and also derived its optimum condition. The dominance of the proposed estimator over the conventional ones has been justified through numerical illustrations carried out over the data set of natural population and population generated through simulation studies using wide varieties of distribution function. Encouraged with the outcomes of the proposed strategy, suitable recommendations to the survey statisticians are also made. [ABSTRACT FROM AUTHOR]

Multiple imputation is a well-established technique for analyzing missing data. Multiple imputed data sets are obtained and analyzed separately using standard complete data techniques. The estimates from the separate analyses are then combined for the purpose of statistical inference. However, the exploratory analysis options of multiple imputed data sets are limited. Biplots are regarded as generalized scatterplots which provide a simultaneous configuration of both samples and variables. A visualization for each of the multiple imputed data sets can be constructed and interpreted individually, but this can become cumbersome and several plots make a unified interpretation challenging. Analogous to multiple imputation, the coordinates of the visualizations can now be regarded as the estimates which are to be pooled in an unbiased manner to construct a final visualization. We propose a GPAbin biplot for a final single visualization after multiple imputation. In a first step, generalized orthogonal Procrustes analysis is used to align the individual biplots before combining their separate coordinate sets into an average coordinate matrix. Finally, this average coordinate matrix is then utilized to construct a single biplot called a GPAbin biplot. A simulation study is used to establish the properties of the final combined GPAbin biplot for varying data characteristics. [ABSTRACT FROM AUTHOR]

To overcome the missing data in sample surveys, the imputation technique plays a significant role in estimating the population mean of the study variable. Clinical data, agricultural data and business data are well-suited examples for this. Keeping this in mind, we have suggested three exponential type imputation methods and the corresponding estimators by making the optimal use of an auxiliary variable. The properties of suggested estimators have been investigated under large sample approximations. The results obtained in this study have been compared with mean estimator, ratio estimator, Kadilar and Cingi (2008) estimators, Diana and Perri (2010) estimators, Al-Omari, Bouza, and Herrera (2013) estimators, and recently Bhushan and Pandey (2016) estimators, Singh et al. (2016) estimators and Singh, Pandey, and Sharma (2020) estimators. It has been shown empirically that the suggested estimators perform effectively over all other estimators. A simulation study is also evaluated in the validation of the present work. [ABSTRACT FROM AUTHOR]

The multiple indicator cluster survey (MICS) is a household survey tool designed to obtain internationally comparable, statistically rigorous data of standardized indicators related to the health situation of children and women. Missing data in a large number of categorical variables are a serious concern for MICS, following complex dependency structures and inconsistency problems that impose severe challenges to the investigators. Despite the popularity of multiple imputation of missing data, its acceptance and application still lag in large-scale studies with complicated data sets such as MICS. We propose interdependent hybrid multiple imputation (HMI) techniques which combines features of existing MI approaches to handle complex missing data in large scale household surveys. The iterative HMI approach is observed to be a good competitor to the existing approaches, with often smaller root mean square errors, empirical standard errors and standard errors. Regardless of any combination, the iterative HMI method is markedly superior to the existing MI methods in terms of computational efficiency. Results from household data example support the capacity of proposed method to handle complex missing data. [ABSTRACT FROM AUTHOR]

Joint modeling technique is a recent advancement in effectively analyzing the longitudinal history of patients with the occurrence of an event of interest attached to it. This procedure is successfully implemented in biomarker studies to examine parents with the occurrence of tumor. One of the typical problem that influences the necessary inference is the presence of missing values in the longitudinal responses as wel l as in covariates. The occurrence of missingness is very common due to the dropout of patients from the study. This article presents an effective and detailed way to handle the missing values in the covariates and response variable. This study discusses the effect of different multiple imputation techniques on the inferences of joint modeling implemented on imputed datasets. A simulation study is carried out to replicate the complex data structures and conveniently perform our analysis to show its efficacy in terms of parameter estimation. This analysis is further illustrated with the longitudinal and survival outcomes of biomarkers' study by assessing proper codes in R programming language. [ABSTRACT FROM AUTHOR]

Missingness or incompleteness of data is a challenge to the survey statisticians in producing the reliable estimates of the desired population parameters. If the missingness pattern is unidentifiable, this is the case of random non-response and to deal with such situations, this work proposes some alternative imputations methods to cope with the missing data. The proposed imputation methods result in some efficient estimation procedures of the current population mean in two-occasion successive sampling. The properties of the resultant estimation procedures have been examined and supplemented with empirical studies. Results have been critically analyzed, and recommendations are made to the survey practitioners. [ABSTRACT FROM AUTHOR]

Missing data are common in medical and social science studies and often face a serious challenge in ultrahigh-dimensional data analysis. In this paper, a nonparametric feature screening approach based on the imputation technique is proposed for ultrahigh-dimensional data with responses missing at random, where the imputation technique is used to replacing each missing value with a set of plausible values. Our approach has many advantages. On one hand, the suggested method relies only on imputation, and its impact is considerably less than that of the feature screening procedure based on the inverse probability weighting approach in missing probabilities. On the other hand, our method does not rely on any model assumption and works generally for all kinds of situations. Simulation studies are conducted to examine the performance of our approach, and a real data example is also presented for illustration. [ABSTRACT FROM AUTHOR]

The missing data often create nuisance to the survey practitioners in producing the reliable estimates of the desired parameters. Keeping this point in view, the present work suggests some alternative imputation techniques to deal with the missing data problem at the beginning of the analysis and proposes some estimation procedures of current population mean in two-occasion successive sampling. The properties of the suggested estimation procedures have been analyzed, and their empirical performances are compared with similar type of estimators designated for whole response situation and another estimator defined for the situation when missing observations observed in the sample data. Based on the fascinating results, suitable recommendations are put forward to the survey practitioners/researchers for their real-life practical applications. [ABSTRACT FROM AUTHOR]

Imputation methods for missing data on a time-dependent variable within time-dependent Cox models are investigated in a simulation study. Quality of life (QoL) assessments were removed from the complete simulated datasets, which have a positive relationship between QoL and disease-free survival (DFS) and delayed chemotherapy and DFS, by missing at random and missing not at random (MNAR) mechanisms. Standard imputation methods were applied before analysis. Method performance was influenced by missing data mechanism, with one exception for simple imputation. The greatest bias occurred under MNAR and large effect sizes. It is important to carefully investigate the missing data mechanism. [ABSTRACT FROM AUTHOR]

Quality of life (QoL) was an important endpoint in the adjuvant breast cancer trials International Breast Cancer Study Group (IBCSG) Trial VI and VII. Here, QoL was considered as a time-dependent effect. The hypothesis explored is that poorer QoL throughout the trial is associated with poorer disease-free survival (DFS) and vice-versa. Potential bias in the parameter estimates is an important concern associated with missing observations. Standard simple and multiple imputation methods were applied to missing QoL assessments before analysis in a time-dependent Cox model. There was no evidence that the patient's QoL is related to the patient's DFS. [ABSTRACT FROM AUTHOR]

It is now a standard practice to replacemissing data in longitudinal surveys with imputed values, but there is still much uncertainty about the best approach to adopt. Using data froma real survey,we compared different strategies combining multiple imputation and the chained equations method, the twomain objectives being (1) to explore the impact of the explanatory variables in the chained regression equations and (2) to study the effect of imputation on causality between successive waves of the survey. Results were very stable fromone simulation to another, and no systematic bias did appear. The critical points of the method lied in the proper choice of covariates and in the respect of the temporal relation between variables. [ABSTRACT FROM AUTHOR]

The multiple imputation technique has proven to be a useful tool in missing data analysis. We propose a Markov chain Monte Carlo method to conduct multiple imputation for incomplete correlated ordinal data using the multivariate probit model. We conduct a thorough simulation study to compare the performance of our proposed method with two available imputation methods – multivariate normal-based and chain equation methods for various missing data scenarios. For illustration, we present an application using the data from the smoking cessation treatment study for low-income community corrections smokers. [ABSTRACT FROM PUBLISHER]

The article focuses mainly on a conditional imputation algorithm of quantile-filling to analyze a new kind of censored data, mixed interval censored and complete data related to interval-censored sample. With the algorithm, the imputed failure times, which are the conditional quantiles, are obtained within the censoring intervals in which some exact failure times are. The algorithm is viable and feasible for the parameter estimation with general distributions, for instance, a case of Weibull distribution that has a moment estimation of closed form by log-transformation. Furthermore, interval-censored sample is a special case of the new censored sample, and the conditional imputation algorithm can also be used to deal with the failure data of interval censored. By comparing the interval-censored data and the new censored data, using the imputation algorithm, in the view of the bias of estimation, we find that the performance of new censored data is better than that of interval censored. [ABSTRACT FROM AUTHOR]

The sensitivity of multiple imputation methods to deviations from their distributional assumptions is investigated using simulations, where the parameters of scientific interest are the coefficients of a linear regression model, and values in predictor variables are missing at random. The performance of a newly proposed imputation method based on generalized additive models for location, scale, and shape (GAMLSS) is investigated. Although imputation methods based on predictive mean matching are virtually unbiased, they suffer from mild to moderate under-coverage, even in the experiment where all variables are jointly normal distributed. The GAMLSS method features better coverage than currently available methods. [ABSTRACT FROM AUTHOR]

Multiple imputation (MI) is now a reference solution for handling missing data. The default method for MI is the Multivariate Normal Imputation (MNI) algorithm that is based on the multivariate normal distribution. In the presence of longitudinal ordinal missing data, where the Gaussian assumption is no longer valid, application of the MNI method is questionable. This simulation study compares the performance of the MNI and ordinal imputation regression model for incomplete longitudinal ordinal data for situations covering various numbers of categories of the ordinal outcome, time occasions, sample sizes, rates of missingness, well-balanced, and skewed data. [ABSTRACT FROM AUTHOR]

There are many methods for analyzing longitudinal ordinal response data with random dropout. These include maximum likelihood (ML), weighted estimating equations (WEEs), and multiple imputations (MI). In this article, using a Markov model where the effect of previous response on the current response is investigated as an ordinal variable, the likelihood is partitioned to simplify the use of existing software. Simulated data, generated to present a three-period longitudinal study with random dropout, are used to compare performance of ML, WEE, and MI methods in terms of standardized bias and coverage probabilities. These estimation methods are applied to a real medical data set. [ABSTRACT FROM AUTHOR]

Sequential regression multiple imputation has emerged as a popular approach for handling incomplete data with complex features. In this approach, imputations for each missing variable are produced based on a regression model using other variables as predictors in a cyclic manner. Normality assumption is frequently imposed for the error distributions in the conditional regression models for continuous variables, despite that it rarely holds in real scenarios. We use a simulation study to investigate the performance of several sequential regression imputation methods when the error distribution is flat or heavy tailed. The methods evaluated include the sequential normal imputation and its several extensions which adjust for non normal error terms. The results show that all methods perform well for estimating the marginal mean and proportion, as well as the regression coefficient when the error distribution is flat or moderately heavy tailed. When the error distribution is strongly heavy tailed, all methods retain their good performances for the mean and the adjusted methods have robust performances for the proportion; but all methods can have poor performances for the regression coefficient because they cannot accommodate the extreme values well. We caution against the mechanical use of sequential regression imputation without model checking and diagnostics. [ABSTRACT FROM AUTHOR]

We propose a new regression-based filter for extracting signals online from multivariate high frequency time series. It separates relevant signals of several variables from noise and (multivariate) outliers. Unlike parallel univariate filters, the new procedure takes into account the local covariance structure between the single time series components. It is based on high-breakdown estimates, which makes it robust against (patches of) outliers in one or several of the components as well as against outliers with respect to the multivariate covariance structure. Moreover, the trade-off problem between bias and variance for the optimal choice of the window width is approached by choosing the size of the window adaptively, depending on the current data situation. Furthermore, we present an advanced algorithm of our filtering procedure that includes the replacement of missing observations in real time. Thus, the new procedure can be applied in online-monitoring practice. Applications to physiological time series from intensive care show the practical effect of the proposed filtering technique. [ABSTRACT FROM AUTHOR]

Although the normality assumption has been regarded as a mathematical convenience for inferential purposes due to its nice distributional properties, there has been a growing interest regarding generalized classes of distributions that span a much broader spectrum in terms of symmetry and peakedness behavior. In this respect, Fleishman's power polynomial method seems to have been gaining popularity in statistical theory and practice because of its flexibility and ease of execution. In this article, we conduct multiple imputation for univariate continuous data under Fleishman polynomials to explore the extent to which this procedure works properly. We also make comparisons with normal imputation models via widely accepted accuracy and precision measures using simulated data that exhibit different distributional features as characterized by competing specifications of the third and fourth moments. Finally, we discuss generalizations to the multivariate case. Multiple imputation under power polynomials that cover most of the feasible area in the skewness-elongation plane appears to have substantial potential of capturing real missing-data trends. [ABSTRACT FROM AUTHOR]

A popular nonparametric treatment of missing value imputation uses methods based on k-nearest neighbors, where the number k of nearest neighbors is fixed without any consideration of the local features of missing values. This article proposes an alternative imputation method based on adaptive nearest neighbors, which takes into account the local features of the data. The proposed method adapts the number of neighbors in imputing the missing values according to the location of the missing values. Efficiency evaluation is then gauged through simulation studies using both simulated and real data. It is shown that the proposed method has distinct advantages over the imputation method based on k-nearest neighbors. [ABSTRACT FROM AUTHOR]

Multiple imputation under the multivariate normality assumption has often been regarded as a viable model-based approach in dealing with incomplete continuous data in the last two decades. A situation where the measurements are taken on a continuous scale with an ultimate interest in dichotomized versions through discipline-specific thresholds is not uncommon in applied research, especially in medical and social sciences. In practice, researchers generally tend to impute missing values for continuous outcomes under a Gaussian imputation model, and then dichotomize them via commonly-accepted cut-off points. An alternative strategy is creating multiply imputed data sets after dichotomization under a log-linear imputation model that uses a saturated multinomial structure. In this work, the performances of the two imputation methods were examined on a fairly wide range of simulated incomplete data sets that exhibit varying distributional characteristics such as skewness and multimodality. Behavior of efficiency and accuracy measures was explored to determine the extent to which the procedures work properly. The conclusion drawn is that dichotomization before carrying out a log-linear imputation should be the preferred approach except for a few special cases. I recommend that researchers use the atypical second strategy whenever the interest centers on binary quantities that are obtained through underlying continuous measurements. A possible explanation is that erratic/idiosyncratic aspects that are not accommodated by a Gaussian model are probably transformed into better-behaving discrete trends in this particular missing-data setting. This premise outweighs the assertion that continuous variables inherently carry more information, leading to a counter-intuitive, but potentially useful result for practitioners. [ABSTRACT FROM AUTHOR]