• 응용통계연구
• /
• 제21권4호
• /
• pp.659-664
• /
• 2008

Missing continuous covariates are pervasive in the use of generalized linear models for medical data. Multiple imputation is the most common and easy-to-do method of dealing with missing covariate data. However, there are always serious warnings in using this method. There should be concern to make imputed values more proper. In this paper, proper imputation from posterior predictive distribution is developed for implementing with arbitrary priors. We use empirical distribution of the posterior for approximating the posterior predictive distribution, to sample from it. This method is preferable in comparison with a presented imputation method of us which uses a full model to impute missing values using available software. The proposed methods are implemented on glucocorticoid data.

• 한국IT서비스학회지
• /
• 제13권4호
• /
• pp.359-375
• /
• 2014

Researchers cannot avoid missing data in collecting data, because some respondents arbitrarily or non-arbitrarily do not answer questions in studies and experiments. Missing data not only increase and distort standard deviations, but also impair the convenience of estimating parameters and the reliability of research results. Despite widespread use of hot deck, researchers have not been interested in it, since it handles missing data in ambiguous ways. Hot deck can be complemented using K-nn, a method of machine learning, which can organize donor groups closest to properties of missing data. Interested in the role of k-nn, this study was conducted to impute missing data based on the hot deck method using k-nn. After setting up imputation of missing data based on hot deck using k-nn as a study objective, deletion of listwise, mean, mode, linear regression, and svm imputation were compared and verified regarding nominal and ratio data types and then, data closest to original values were obtained reasonably. Simulations using different neighboring numbers and the distance measuring method were carried out and better performance of k-nn was accomplished. In this study, imputation of hot deck was re-discovered which has failed to attract the attention of researchers. As a result, this study shall be able to help select non-parametric methods which are less likely to be affected by the structure of missing data and its causes.

• Journal of the Korean Data and Information Science Society
• /
• 제26권4호
• /
• pp.971-984
• /
• 2015

Imputation procedures fill-in missing values, thereby enabling complete data analyses. Fully efficient fractional imputation (FEFI) and multiple imputation (MI) create multiple versions of the missing observations, thereby reflecting uncertainty about their true values. Methods have been described for hypothesis testing with multiple imputation. Fractional imputation assigns weights to the observed data to compensate for missing values. The focus of this article is the development of tests of independence using FEFI for partially classified two-way contingency tables. Wald and deviance tests of independence under FEFI are proposed. Simulations are used to compare type I error rates and Power. The partially observed marginal information is useful for estimating the joint distribution of cell probabilities, but it is not useful for testing association. FEFI compares favorably to other methods in simulations.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제32권2호
• /
• pp.99-107
• /
• 2005

본 논문에서는 단일염기변이(SNP: Single Nucleotide Polymorphism)와 같은 유전형(Rcnotype)자료에서 결측치가 발생하였을 경우 유전형 자료의 특이성을 고려해 자료 원래의 정보손실을 최소화하는 대치법인 연관불균형 기반의 대치법(linkage disequilibrium- based imputation)과 일배체형 기반의 대치법(haplotype-based imputation)을 제시한다. 이러한 결측치 대치는 실험상에서 발생하는 결측치에 의한 중요한 정보의 손실을 최소화 한다는 점에서 필요한 방법이다. 일반적으로 그동안 생물학 자료의 결측치 대치는 대부분 주형질 대치법(major allele imputation)이 활용되어왔는데 유전형 자료에서의 이 방법의 사용은 사료의 특이성으로 인하여 결측치에 대한 높은 오차율(error rate)을 보임으로서 자료의 신뢰성을 떨어뜨릴 수 있다. 본 논문에서는 유전형 자료인 단일염기변이 자료의 시뮬레이션을 통하여 기존의 주형질 대치법과 논문에서 제안된 연관불균형 기반의 대치법과 일배체형 기반의 대치법을 비교하고 그 결과를 보여 준다.

• 한국통계학회:학술대회논문집
• /
• 한국통계학회 2003년도 춘계 학술발표회 논문집
• /
• pp.237-242
• /
• 2003

Various kinds of estimation methods have been developed for imputation of categorical missing data. They include modal category method, logistic regression, and association rule. In this study, we propose two imputation methods (neural network fusion and voting fusion) that combine the results of individual imputation methods. A Monte-Carlo simulation is used to compare the performance of these methods. Five factors used to simulate the missing data are (1) true model for the data, (2) data size, (3) noise size (4) percentage of missing data, and (5) missing pattern. Overall, neural network fusion performed the best while voting fusion is better than the individual imputation methods, although it was inferior to the neural network fusion. Result of an additional real data analysis confirms the simulation result.

• 한국컴퓨터정보학회논문지
• /
• 제29권4호
• /
• pp.39-46
• /
• 2024

많은 연구자들이 다양한 모델을 이용하여 물의 수질을 평가하기 위해 노력하고 있다. 평가 모델에는 결측값이 없는 데이터셋이 필요하지만, 관측 데이터셋에는 결측값이 다수 포함되는 것이 현실이다. 단순히 결측값을 삭제하는 방법은 경우에 따라 기저 데이터의 분포를 왜곡시키고 모델의 예측성능에도 편의(bias)를 불러올 위험성이 있다. 본 연구에서는 수질 데이터의 결측값 처리에 적합한 기법을 탐색하기 위해, 기존의 KNN과 MICE Imputation, 그리고 생성형 신경망 모델인 Autoencoder와 Denoising Autoencoder를 기반으로 몇 가지 대치 기법을 실험하였다. 실험 결과, KNN과 MICE Imputation의 결과를 평균한 Combined Imputation이 실측치에 가장 가깝게 값을 추정하였으며, 이 기법을 적용하여 결측값을 처리한 관측 데이터셋을 support vector machine과 ensemble 기반의 분류 모델로 평가한 결과, 결측값을 삭제했을 때에 비해 Accuracy, F1 score, ROC-AUC score, 그리고 MCC(Mathews Correlation Coefficient) 지표가 향상되었다.

• Journal of information and communication convergence engineering
• /
• 제21권1호
• /
• pp.62-67
• /
• 2023

This paper proposes an imputation method using a bidirectional k-nearest components search based local linear regression method. The bidirectional k-nearest-components search method selects components in the dynamic range from the missing points. Unlike the existing methods, which use a fixed-size window, the proposed method can flexibly select adjacent components in an imputation problem. The weight values assigned to the components around the missing points are calculated using local linear regression. The local linear regression method is free from the rank problem in a matrix of dependent variables. In addition, it can calculate the weight values that reflect the data flow in a specific environment, such as a blackout. The original missing values were estimated from a linear combination of the components and their weights. Finally, the estimated value imputes the missing values. In the experimental results, the proposed method outperformed the existing methods when the error between the original data and imputation data was measured using MAE and RMSE.

• Communications for Statistical Applications and Methods
• /
• 제26권3호
• /
• pp.325-336
• /
• 2019

In this article we extend the discrete Weibull regression model in the presence of missing data. Discrete Weibull regression models can be adapted to various type of dispersion data however, it is not widely used. Recently Yoo (Journal of the Korean Data and Information Science Society, 30, 11-22, 2019) adapted the discrete Weibull regression model using single imputation. We extend their studies by using multiple imputation also with several various settings and compare the results. The purpose of this study is to address the merit of using multiple imputation in the presence of missing data in discrete count data. We analyzed the seventh Korean National Health and Nutrition Examination Survey (KNHANES VII), from 2016 to assess the factors influencing the variable, 1 month hospital stay, and we compared the results using discrete Weibull regression model with those of Poisson, negative Binomial and zero-inflated Poisson regression models, which are widely used in count data analyses. The results showed that the discrete Weibull regression model using multiple imputation provided the best fit. We also performed simulation studies to show the accuracy of the discrete Weibull regression using multiple imputation given both under- and over-dispersed distribution, as well as varying missing rates and sample size. Sensitivity analysis showed the influence of mis-specification and the robustness of the discrete Weibull model. Using imputation with discrete Weibull regression to analyze discrete data will increase explanatory power and is widely applicable to various types of dispersion data with a unified model.

• 응용통계연구
• /
• 제16권2호
• /
• pp.407-426
• /
• 2003

본 연구에서는 범주형 반복측정자료의 일반화추정방정식(GEE)모형에서 결측이 발생할 경우 결측값 대체(imputation)방법들에 대한 성능을 비교하고자 한다. 설명변수 X가 부분적으로 결측을 갖는 경우 GEE추정량을 계산할 수 없다. 본 논문에서는 시점에 따라 값이 변하는 설명변수에 결측이 있는 경우 GEE모형에서 결측값을 추정하는 7가지의 대체방법을 다루며, 실제자료와 모의실험을 통하여 대체방법별 GEE추정량의 성질을 연구한다. 대체방법별 GEE추정량의 성능을 비교하기 위해 우리는 반응변수가 범주형인 반복측정모형에서 완전자료의 GEE추정량과 완전자료에서 결측을 생성하여 결측값에 각 대체방법을 적용하여 대체한 후 구한 GEE추정량을 비교한다. 대체방법으로는 (1) 단순삭제 (2) 표본 평균대체 (3) 행 평균대체 (4) 횡 시점 회귀대체 (5) 이월대체 (6) 베이지안 붓스트랩 (7) 근사적 베이지안 붓스트랩에 대해서 살펴본다. 결측과정(missing mechanism)은 무시할 수 있는 무응답(ignorable nonresponse)을 가정하며, 결측 발생에 대해서는 원자료의 시점 무응답 패턴(wave nonresponse pattern)을 고려하여 발생시키거나 또는 시점 무응답 패턴을 고려하지 않고 단순임의추출로 결측을 발생시키는 방법을 각각 고려한다.

• Communications for Statistical Applications and Methods
• /
• 제26권4호
• /
• pp.411-430
• /
• 2019

Tumor development is driven by complex combinations of biological elements. Recent advances suggest that molecularly distinct subtypes of breast cancers may respond differently to pathway-targeted therapies. Thus, it is important to dissect pathway disturbances by integrating multiple molecular profiles, such as genetic, genomic and epigenomic data. However, missing data are often present in the -omic profiles of interest. Motivated by genomic data integration and imputation, we present a new statistical framework for pathway significance analysis. Specifically, we develop a new strategy for imputation of missing data in large-scale genomic studies, which adapts low-rank, structured matrix completion. Our iterative strategy enables us to impute missing data in complex configurations across multiple data platforms. In turn, we perform large-scale pathway analysis integrating gene expression, copy number, and methylation data. The advantages of the proposed statistical framework are demonstrated through simulations and real applications to breast cancer subtypes. We demonstrate superior power to identify pathway disturbances, compared with other imputation strategies. We also identify differential pathway activity across different breast tumor subtypes.