결측이 있는 이산형 공변량에 대한 Cox비례위험모형의 패턴-혼합 모델

육태미,송주원,Youk, Tae-Mi,Song, Ju-Won 한국통계학회 2012 응용통계연구 Vol.25 No.2

공변량에 결측이 발생한 Cox 비례위험 모형을 적합할 때, 결측이 발생하는 개체를 모두 제거한 후 분석을 실시한다면 정보 손실에 의해 비효율적이고 결측의 발생 메커니즘이 완전 임의 결측(missing completely at random; MCAR)이 아니라면 모수의 추정값에 편향이 발생할 수 있다. Cox 비례위험 회귀모형의 공변량에 결측이 있는 경우 적용할 수 있는 여러 가지 방법들이 제안되어져 왔으나 이 분석들은 선택모델(selection model)에 기반하고 있다. 본 연구에서는 Little (1993)이 제안한 패턴-혼합 모델(pattern-mixture model)을 사용하여 Cox 비례위험 회귀모형에서 생존시간과 결측 메커니즘의 결합분포를 모델화 하고, 여러 가지 제약에 근거한 생존 분석의 결과를 비교하였다. 모의실험을 통해서 패턴-혼합 모델의 제약(restrictions)에 따른 모수 추정의 민감도를 확인하였고 결측을 무시한 채 분석한 결과 및 선택모형에 근거한 분석결과와 비교하였다. 패턴-혼합 모델의 제약에 따라 공변량의 결측으로 인한 모수 추정의 민감성 정도를 쥐백혈병 자료 예제를 통해 설명하였다. When fitting a Cox proportional hazards model with missing covariates, it is inefficient to exclude observations with missing values in the analysis. Furthermore, if the missing-data mechanism is not Missing Completely At Random(MCAR), it may lead to biased parameter estimation. Many approaches have been suggested to handle the Cox proportional hazards model when covariates are sometimes missing, but they are based on the selection model. This paper suggest an approach to handle Cox proportional hazards model with missing covariates by using the pattern-mixture model (Little, 1993). The pattern-mixture model is expressed by the joint distribution of survival time and the missing-data mechanism. In the pattern-mixture model, many models can be considered by setting up various restrictions, and different results under various restrictions indicate the sensitivity of the model due to missing covariates. A simulation study was conducted to show the sensitivity of parameter estimation under different restrictions in a pattern-mixture model. The proposed approach was also applied to mouse leukemia data.

개입모형을 이용한 서울 아파트매매거래량 분석

육태미(Youk, Tae Mi),이수민(Lee, Soo Min) 한국부동산원 2018 부동산분석 Vol.4 No.1

시장 경제를 반영하는 여러 가지 지표들은 외부의 상황에 따라 증가, 감소를 반복하는 시계열을 보인다. 이때 정책 변화나 뉴스 보도 등의 사건이 시계열자료에 영향을 미칠 수 있는 경우 이를 개입(intervention)이라고 하며, 실제 자료 분석 시에는 이를 반영해 주는 모형이 필요하다. 본 연구에서는 2017년 하반기에 발표된 부동산 정책들의 변화가 서울 아파트매매거래량에 미치는 영향을 개입효과를 고려한 계절형 ARIMA 모형을 이용하여 분석하였다. 분석 결과 2017년 6월 이후 3개월간 지속 발표된 부동산 대책은 순간적으로 서울 아파트매매거래량을 감소시키는 효과를 보였지만 그 지속 효과는 거의 없었던 것으로 나타났다. 기존의 보고서들에서 다루고 있는 정책 변화 전 · 후 차이의 단순 빈도 비교보다 좀 더 이론적인 접근을 가능하게 하는데에 본 연구의 의의가 있다. 또한 시계열에 영향을 주는 변수를 구분하고 그로 인한 변화 정도를 추정하는 것은 향후 안정적인 정책을 수립하는데 기여할 수 있으리라 본다. Various indicators reflecting the market economy show a time series that repeatedly increases and decreases depending on external situation. If an event, such as policy change and news report, can affect time series data, it is called intervention, and in actual data analysis, a model that reflects it is required. In this study, we analyze the effect of real estate policies announced in the second half of 2017 on the change of Seoul apartment trading volume, using a seasonal ARIMA model that reflects intervention effects. The results show that real estate policies, which were announced consecutively for a period of three months from June 2017, reduced the Seoul apartment trading volume momentarily, but had no long-lasting effect. This study has significant implications in that it allows a more theoretical approach than the simple comparison of differences between pre and post the change of policies in previous studies. In addition, it is presumed that the identification of variables affecting the time series and the estimation of the degrees of changes caused thereby will contribute to establishing a stable policy in the future.

매월 조사되는 주택 가격 변동률의 이상치 탐색 방법에 관한 연구

육태미 ( Tae Mi Youk ),방송희 ( Song Hee Bang ),이재성 ( Jae Sung Lee ) 한국부동산분석학회 2013 不動産學硏究 Vol.19 No.4

In statistical theory, an outlier is a value that is numerically distant from overall pattern of a distribution. It may be a meaningful observation, but it comes from an error in survey, data entry and process in most cases. Detection and handling are needed because outliers by errors debase the statistical quality leading to biased parameter estimation. Generally, traditional Box-plot or Z-score are very useful for univariate outlier detection and a Median rule could be applied in the non-Gaussian case. These methods calculate the tolerance interval that defines the range of acceptable observation values. Outlier detection for periodic surveys would consider the past view, because it is based on a ratio of value comparing the current time with previous time. If time period, however, is short, a state to get many unchanged values can occur. In this case, the ratio is centered at 1, and therefore outlier detection method reflecting this factor is required. This paper considers Quartile Method with power transformation and Hidiroglou-Berthelot(1986) method that is efficient in periodic data. The methods were applied to housing sales price. We suggest an outlier detection method for real-world data. In addition, we also analyzed data using Tukey Algorithm of United Kingdom``s office of National Statistic(ONS).

2008년 안산시 지역정신건강 조사

최현희,김지선,육태미 고려대학교 대학원 통계학과 2008 통계상담 Vol.사례 21 No.2

현대인들은 과도한 스트레스, 무력감, 불안감을 느끼며 정신건강에 위협을 받고 있다. 이러한 시점에서 지역주민과 정신장애인 및 가족, 관련 전문직의 조사를 통해 정신장애에 대한 인식을 알아보고, 그동안의 정신보건 서비스에 대한 만족도와 요구도 정도에 대하여 살펴보고자 한다.

Abrupt Increase in Rate of Imipenem Resistance in Acinetobacter baumannii Complex Strains Isolated from General Hospitals in Korea and Correlation With Carbapenem Administration During 2002–2013

김영아,박윤수,육태미,이혁민,이경원 대한진단검사의학회 2018 Annals of Laboratory Medicine Vol.38 No.2

Dear Editor, Acinetobacter baumannii is a cause of healthcare-associated infections and one of the most important pathogens found in intensive care units (ICUs). The recent dramatic spread of carbapenem-resistant A. baumannii complex (CRAB) in Korea has raised substantial concern in medical practice owing to the lack of adequate therapeutic options [1, 2]. It is well-known that the selective pressure of antibiotics contributes to emergence of new resistant strains [3]. However, there is no sufficient evidence for the association between carbapenem administration and emergence of imipenem-resistant A

Incidence and Clinical Outcomes of Febrile Neutropenia in Adult Cancer Patients with Chemotherapy Using Korean Nationwide Health Insurance Database

김달용,이수현,육태미,홍수정 연세대학교의과대학 2021 Yonsei medical journal Vol.62 No.6

Purpose: The aim of this study was to evaluate the episodes of febrile neutropenia (FN) in patients with gastric cancer (GC),colorectal cancer (CRC), lung cancer (LC), and breast cancer (BC); and to identify the incidence and trends of admission rates, aswell as factors affecting mortality. Materials and Methods: Using nationwide claims data, all new admissions to hospitals for FN were selected. We evaluated theincidence of FN and mortality-related clinical factors in adult cancer patients who received cytotoxic chemotherapy from January2004 to December 2013. Results: While the incidence of FN increased, the length of hospitalization decreased in Korea. The incidence of FN was 19.8% inLC patients, 15.5% in GC patients, 13.3% in BC patients, and 9.5% in CRC patients. The overall in-hospital mortality of FN was12.9% and showed a decreasing trend. Admission rates to intensive care units and in-hospital mortality were the highest for lungcancer (15.2% and 19.3%, respectively). Age and sepsis syndrome were risk factors for in-hospital mortality for all cancer types. Conclusion: Careful observation and active prophylaxis should be considered for patients at high risk of FN.

Correlation of Aminoglycoside Consumption and Amikacin- or Gentamicin-Resistant Pseudomonas aeruginosa in Long-Term Nationwide Analysis: Is Antibiotic Cycling an Effective Policy for Reducing Antimicrobial Resistance?

김영아,박윤수,육태미,이혁민,이경원 대한진단검사의학회 2018 Annals of Laboratory Medicine Vol.38 No.2

Dear Editor, Bacteria continuously develop, acquire, and spread numerous resistance patterns shortly after new antibiotics reach the market. Recently, carbapenem-resistant Enterobacteriaceae (CRE) and colistin-resistant Acinetobacter baumannii are emerging, which may lead us into the pre-antibiotic era. Pipelines of new antibiotics are becoming thinner and thinner across all therapeutic areas due to difficulties in innovation and challenges of regulatory hurdles. Under such a situation, maintaining antibiotic effectiveness in the long term requires not only innovation to develop new antibiotics, but also conservation of the effectiveness of existing antibiotics.

Incidence of Cerebral Palsy in Korea and the Effect of Socioeconomic Status: A Population-Based Nationwide Study

김성우,전하라,신지철,육태미,김지용 연세대학교의과대학 2018 Yonsei medical journal Vol.59 No.6

Purpose: Cerebral palsy (CP) is a neurodevelopmental disorder that causes serious disability. Prematurity and low birth weight(LBW) are known to be the strongest risk factors of CP. While socioeconomic status (SES) has been found to influence the occurrenceof CP, prematurity, and LBW, no studies have investigated this effect in Korea. The aim of this study was to evaluate the incidenceof CP, prematurity, and LBW in Korea, as well as the effect of SES thereon. Materials and Methods: Data were obtained from the National Health Information Database from 2007 to 2013; persons with ahistory of CP, prematurity, and LBW were investigated by year. SES was defined in accordance with income quintiles, birth regions,and coverage classification. Results: The incidence of CP decreased over the last five years, despite increased rates of prematurity and LBW. CP incidence wassignificantly lower in affluent groups than in the most deprived group, although this difference disappeared after controlling forconfounders. The incidence of CP was significantly higher in medical aid beneficiaries, even after controlling for confounders. Conclusion: CP incidence in Korea has decreased over the last five years, despite an increase in high-risk deliveries. Income levelhad no effect in CP incidence. These results may aid CP management and prevention policies.

Gradual Increase in Hematologic Malignancy in Korea from 2005 to 2015 Based on the National Health Insurance Service Data

한유진,김영진,김민진,변자민,육태미,윤회수,이재희,이우인,박태성,유종하 대한진단검사의학회 2020 Laboratory Medicine Online Vol.10 No.2

Background: Hematologic malignancies have a relatively lower prevalence than major solid cancers, although the incidence of hematologic malignancies has significantly increased in recent years. However, understanding the current status of hematologic malignancy is significantly challenging because basic data regarding this malignancy are insufficient in the Korean population. Methods: From 2005 to 2015, the status of seven codes of hematologic malignancy, containing 24 subcodes defined using a classification defined by the Korean Classification of Disease-6, was analyzed. The number of new patients, crude incidence rate, prevalence rate, and age-standardized incidence rate were also investigated. Results were analyzed based on National Health Insurance Service (NHIS) data. Results: The number of new patients showed an overall increase over time and a rate of increase up to 56.7% for 10 years. The number of male patients was higher than that of female patients, with the majority of patients aged greater than 60 years. The incidence and prevalence rates have increased steadily. Conclusions: Consistent with the previous studies, this study might be useful to understand the current status of hematologic malignancy and might contribute to the improvement of national public healthcare.

Disability Registration State of Children With Cerebral Palsy in Korea

김성우,전하라,김윤석,최수진,육태미,김지용 대한재활의학회 2018 Annals of Rehabilitation Medicine Vol.42 No.5

Objective To investigate the disability registration state of children with cerebral palsy (CP) in Korea. Methods Based on the National Health Information Database, the disability registration state was examined for brain lesion disability and other possible complicated disabilities accompanying brain disorder in children diagnosed with CP aged up to 5 years old who were born between 2002 and 2008.Results Of children diagnosed with CP, 73.1% were registered as having brain lesion disability for the first time before they turned 2 years old. The younger the children, the more likely they will have 1st and 2nd degree disability. However, when the age of children is increased, such likelihood is decreased. The percentage of children registered as having overlapping disabilities was 7%–20%. Conclusion It is important to establish a more accurate standard to rate disability and provide national support systems for children with CP with various severities and multiple disabilities. By reorganizing the current disability registration system for pediatric brain lesions, the system could serve as a classification standard to provide medical and social welfare services.