• Journal of the Korean Data and Information Science Society
• /
• 제16권1호
• /
• pp.95-106
• /
• 2005

Complete-case analysis is easy to carry out and it may be fine with small amount of missing data. However, this method is not recommended in general because the estimates are usually biased and not efficient. There are numerous alternatives to complete-case analysis. A natural alternative procedure is available-case analysis. Available-case analysis uses all cases that contain the variables required for a specific task. The EM algorithm is a general approach for computing maximum likelihood estimates of parameters from incomplete data. These methods and multiple imputation(MI) are reviewed and the performances are compared by simulation studies in monotone missing pattern.

• 한국전기전자재료학회:학술대회논문집
• /
• 한국전기전자재료학회 2004년도 추계학술대회 논문집 Vol.17
• /
• pp.327-330
• /
• 2004

Because of its effectiveness for the PD(partial discharge) pattern recognition, PSA(Pulse Sequence Analysis) has been considered as a new analytic method instead of conventional PRPDA(Phase Resolved Partial Discharge Analysis). However, PSA has a big problem that can misanalyze patterns in case of data missing resulting from poor sensitivity because it analyses the correlation between sequential pulses, which leads to hesitate to apply it to on-site. Therefore, in this paper, the problems of PSA such as data missing and noise adding cases were investigated. For the purpose, PD data obtained from various defects including noise adding data were used and analysed, The result showed that both cases can cause fatal errors in recognizing PD patterns. In case of the data missing, the error depends on the kinds of defect and the degree of degradation. Also, it could be noticed that the error due to adding noises was larger than that due to some data missing.

• 한국전기전자재료학회논문지
• /
• 제18권5호
• /
• pp.484-489
• /
• 2005

Because of its effectiveness for the PD(Partial Discharge) pattern recognition, PSA(Pulse Sequence Analysis) has been considered as a new analytic method instead of conventional PRPDA(Phase Resolved Partial Discharge Analysis). However, it is generally thought that PSA has some possibility to misjudge patterns in case of data-missing resulting from poor sensitivity because it analyses the correlation between sequential pulses, which leads to hesitate to apply it to on-site. Therefore, in this paper, the problems of PSA such as data-missing and noise-adding cases were investigated. for the purpose, PD data obtained from various defects including noise-adding data were used and analyzed. As a result, it was shown that both cases could cause fatal errors in recognizing PD patterns. In case of the data missing, the error was dependant on the kinds of defect and the degree of degradation Also, it could be noticed that the error due to adding noises was larger than that due to some data missing.

• 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.

• 인터넷정보학회논문지
• /
• 제11권2호
• /
• pp.129-142
• /
• 2010

무선선서네트워크의 주요 응용분야 중 하나가 유비쿼터스 헬스케어 시스템이다. 하지만 무선센서네트워크가 가지고 있는 과제중의 하나는 데이터 중에 나타나는 높은 손실 율이다. 바이오 센서로부터 들어오는 데이터는 기지국에 도착되지 않을 수 있으며, 이 값은 손실 값(missing value)이 된다. 본 논문은 기지국에서 데이터를 수집하고, 손실 값을 처리한 후, 증상 패턴에 따라 건강상태를 분류하여, 비상시에 적절한 행동을 취할 수 있도록 하는 헬스케어 모니터 에이전트(HMA)를 제안한다. 이 에이전트는 유비쿼터스 헬스케어 환경에 적용되며, 건강상태를 인지하기 위한 증상패턴으로 바이오 센서 및 환자의 가족력으로 부터 생성된 데이터를 사용한다. 손실 값이 나타나면 HMA는 분류하기 전에 증상패턴의 손실 값을 채우기 위한 예측 알고리즘을 수행한다. 시뮬레이션 결과 HMA를 사용한 예측알고리즘이 다른 방법들에 비해 더 정확하게 증상패턴을 분류함을 보여주었다.

• Journal of the Korean Data and Information Science Society
• /
• 제20권4호
• /
• pp.757-764
• /
• 2009

In discriminant analysis, we consider a special pattern which contains a block of missing observations. We assume that the two populations are equally likely and the costs of misclassification are equal. In this situation, we consider the bootstrap confidence intervals of the error rate in the circular models when the covariance matrices are equal and not equal.

• Journal of information and communication convergence engineering
• /
• 제16권4호
• /
• pp.242-247
• /
• 2018

The data of K-Water waterworks is collected from various sensors and used as basic data for the operation and analysis of various devices. In this way, the importance of the sensor data is very high, but it contains misleading data due to the characteristics of the sensor in the external environment. However, the cleansing method for the missing data is concentrated on the prediction of the missing data, so the research on the detection and prediction method of the missing data is poor. This is a study to detect wrong data by converting collected data into quintiles and patterning them. It is confirmed that the accuracy of detecting false data intentionally generated from real data is higher than that of the conventional method in all cases. Future research we will prove the proposed system's efficiency and accuracy in various environments.

• Journal of the Korean Data and Information Science Society
• /
• 제15권4호
• /
• pp.945-955
• /
• 2004

'Complete-case analysis' is easy to carry out and it may be fine with small amount of missing data. However, this method is not recommended in general because the estimates are usually biased and not efficient. There are numerous alternatives to complete-case analysis. One alternative is the single imputation. Some of the most common single imputation methods are reviewed and the performances are compared by simulation studies.

• Asian Pacific Journal of Cancer Prevention
• /
• 제17권10호
• /
• pp.4567-4575
• /
• 2016

A number of studies have evidenced the effect of modifiable lifestyle factors such as diet, breastfeeding and nutritional status on breast cancer risk. However, none have addressed the missing data problem in nutritional epidemiologic research in South America. Missing data is a frequent problem in breast cancer studies and epidemiological settings in general. Estimates of effect obtained from these studies may be biased, if no appropriate method for handling missing data is applied. We performed Multiple Imputation for missing values on covariates in a breast cancer case-control study of $C{\acute{o}}rdoba$ (Argentina) to optimize risk estimates. Data was obtained from a breast cancer case control study from 2008 to 2015 (318 cases, 526 controls). Complete case analysis and multiple imputation using chained equations were the methods applied to estimate the effects of a Traditional dietary pattern and other recognized factors associated with breast cancer. Physical activity and socioeconomic status were imputed. Logistic regression models were performed. When complete case analysis was performed only 31% of women were considered. Although a positive association of Traditional dietary pattern and breast cancer was observed from both approaches (complete case analysis OR=1.3, 95%CI=1.0-1.7; multiple imputation OR=1.4, 95%CI=1.2-1.7), effects of other covariates, like BMI and breastfeeding, were only identified when multiple imputation was considered. A Traditional dietary pattern, BMI and breastfeeding are associated with the occurrence of breast cancer in this Argentinean population when multiple imputation is appropriately performed. Multiple Imputation is suggested in Latin America's epidemiologic studies to optimize effect estimates in the future.

• 대한교통학회지
• /
• 제26권2호
• /
• pp.109-119
• /
• 2008

대중교통 교통카드 도입 이후, 점차 이용율이 증가되고 있다. 카드 데이터를 통해 얻을 수 있는 자료를 고려할 때 대중교통 카드 이용의 증가는 통행패턴 분석 및 정책적 측면에서 중요한 의미를 가지고 있다. 그 중에서 특히 죤별 대중교통 통행수요(O/D)를 손쉽게 파악할 수 있다는 점에서 높은 중요성을 가진다. 카드데이터를 통해 대중교통 죤별 통행수요(O/D)를 파악함에 있어서 데이터 자체의 오류에 대한 분석이나 결측에 대한 보완 과정이 반드시 필요하다. 본 연구에서는 반드시 선행되어야 할 과제이지만 아직 연구사례가 없었던 카드데이터의 오류와 결측에 관해 살펴보았다. 그 결과, 통행수요(O/D)분석과 관련한 오류나 결측에 대한 특성을 제시하였고, 결측에 대한 보정방안을 제안하였다. 그리고 제시된 결측방안들에 대한 적용 및 평가와 함께 활용방안을 제시하여, 향후 보다 신뢰성있는 대중교통 OD구축을 위한 기반을 마련하였다.