붓스트랩 방법을 적용한 확률계수 자기회귀 모형에 대한 로버스트 구간추정

조나래,임도상,이성덕 한국통계학회 2019 응용통계연구 Vol.32 No.1

We compared the confidence intervals of estimators using various bootstrap methods for a Random Coefficient Autoregressive(RCA) model. We consider a Quasi score estimator and M-Quasi score estimator using Huber, Tukey, Andrew and Hempel functions as bounded functions, that do not have required assumption of distribution. A standard bootstrap method, percentile bootstrap method, studentized bootstrap method and hybrid bootstrap method were proposed for the estimations, respectively. In a simulation study, we compared the asymptotic confidence intervals of the Quasi score and M-Quasi score estimator with the bootstrap confidence intervals using the four bootstrap methods when the underlying distribution of the error term of the RCA model follows the normal distribution, the contaminated normal distribution and the double exponential distribution, respectively. 비선형 시계열인 확률계수 자기회귀(random coefficient autoregressive; RCA) 모형에 대하여 여러 가지 방법을 이용한 추정량의 신뢰구간 비교하였다. RCA 모형에 대하여 자료의 분포를 가정하지 않아도 되는 Quasi 스코어 추정량과 Huber, Tukey, Andrew, Hempal 4가지 유계함수를 이용한 M-Quasi 스코어 추정량을 제시하였다. 이러한 추정량에 대하여 표준 붓스트랩 방법, 백분위수 붓스트랩 방법, 스튜던트화 붓스트랩 방법, 하이브리드 붓스트랩 방법을 이용한 신뢰구간을 구하였다. 모의실험을 통하여 RCA 모형의 오차항의 분포가 정규분포, 오염정규분포, 이중지수분포를 따를 때 Quasi 스코어 추정량과 M-Quasi 스코어 추정량들의 근사적 신뢰구간과 네가지 붓스트랩 방법을 이용한 신뢰구간을 비교하였다.

지역 간 건강수준 비교를 위한 표준화율 적용의 적절성 평가: 2008년 지역사회건강조사를 바탕으로

권근용,임도상,박은자,정지선,강기원,김윤아,김호,조성일,Kwon, Geun-Yong,Lim, Do-Sang,Park, Eun-Ja,Jung, Ji-Sun,Kang, Ki-Won,Kim, Yun-A,Kim, Ho,Cho, Seong-Il 대한예방의학회 2010 예방의학회지 Vol.43 No.2

Objectives: This study shows the issues that should be considered when applying standardized rates using Community Health Survey(CHS) data. Methods: We analyzed 2008 CHS data. In order to obtain the reliability of standardized rates, we calculated z-score and rank correlation coefficients between direct standardized rate and indirect standardized rate for 31 major indices. Especially, we assessed the change of correlations according to population composition (age and sex), and characteristics of the index. We used Mantel-Haenszel chi-square to quantify the difference of population composition. Results: Among 31 major indices, 29 indices' z-score and rank correlation coefficients were over 0.9. However, regions with larger differences in population composition showed lower reliability. Low reliability was also observed for the indices specific to subgroups with small denominator such as 'permanent lesion from stroke', and the index with large regional variations in age-related differences such as 'obtaining health examinations'. Conclusions: Standardized rates may have low reliability, if comparison is made between areas with extremely large differences in population composition, or for indicies with large regional variations in age-related differences. Therefore, the special features of standardized rates should be considered when health state are compared among areas.

자발적협약제도 시행효과 계량에 관한 연구

조기선(Ki-Seon Cho),임도상(Do-Sang Lim),이창호(Chang-Ho Rhee) 대한전기학회 2006 대한전기학회 학술대회 논문집 Vol.2006 No.7

본 논문은 에너지절약 및 온실가스 배출감소를 목적으로 시행하고 있는 에너지절약정책 중 산업체 자발적협약(Voluntary Agreement)제도의 정책효과를 계량화하기 위한 접근방법을 제시한다. 시책 추진효과의 계량은 2단계 접근으로, 1단계에서 절감효과(에너지 절감 및 온실가스 저감)를 파악하고, 2단계에서 이를 가치화하는 과정으로 수행된다.

지역사회 건강조사자료에 대한 다수준 모형의 활용성에 대한 연구

이지혜,이두리,김선우,임도상,허태영 한국자료분석학회 2013 Journal of the Korean Data Analysis Society Vol.15 No.3

In this study, we propose the procedure and methods of multilevel analysis using the Community Health Survey data from Korea Centers for Disease Control and Prevention and analyze the effect of various health status on the quality of life. The variables that affect the quality of life as the dependent variable are subjective health status, oral health and stress. In order to determine the appropriate level of analysis, r_wg (within group agreement), ICC(1), ICC(2), WABA (within and between analysis) etc. widely used in multilevel analysis was used as the evaluation measure. As a result of using evaluation measures to determine the level of analysis, single level model is more appropriate than multilevel model. Also, the result of multilevel analysis is that the influence of group level are not statistically significant. Therefore, we conclude that the evaluation measure used to determine the level of analysis is appropriate. 본 연구에서는 질병관리본부에서 매년 조사하고 있는 지역사회 건강 조사자료를 이용하여 충청북도 지역을 대상으로 국민의 삶의 질에 미치는 영향을 다수준 분석의 절차와 방법을 이용하여 분석하였다. 종속 변수인 삶의 질에 영향을 미치는 변수로는 주관적 건강상태, 구강건강, 스트레스를 이용하였으며 지역사회 건강조사자료의 다수준 분석의 활용성을 확인하고 적합한 분석의 수준을 결정하기 위해 다수준 분석에서 많이 사용되고 있는 집단내 일치도 지수(r_wg), 급내상관계수(ICC(1)과 ICC(2)), WABA(within and between analysis) 등을 평가 측도로 사용하였다. 분석의 수준을 결정하기 위해 평가 측도를 사용한 결과 다수준 모형보다는 단일수준 모형이 적합함을 보였으며 다수준 분석 결과 또한 집단 수준의 영향력이 통계적으로 유의하지 않아 분석 수준 결정에 사용된 평가 측도가 적절함을 알 수 있었다.

Household secondary attack rates and risk factors during periods of SARS-CoV-2 Delta and Omicron variant predominance in the Republic of Korea

Jin Lee,Mijeong Ko,Seontae Kim,임도상,Park Gemma,이상은 질병관리본부 2023 Osong Public Health and Research Persptectives Vol.14 No.4

Objectives: The household secondary attack rate (SAR) of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is an important indicator for community transmission. This study aimed to characterize transmission by comparing household SARs and identifying risk factors during the periods of Delta and Omicron variant predominance in Republic of Korea.Methods: We defined the period of Delta variant predominance (Delta period) as July 25, 2021 to January 15, 2022, and the period of Omicron variant predominance (Omicron period) as February 7 to September 3, 2022. The number of index cases included was 214,229 for the Delta period and 5,521,393 for the Omicron period. To identify the household SARs and risk factors for each period, logistic regression was performed to determine the adjusted odds ratio (aOR).Results: The SAR was 35.2% for the Delta period and 43.1% for the Omicron period. The aOR of infection was higher in 2 groups, those aged 0 to 18 years and ≥75 years, compared to those aged 19 to 49 years. Unvaccinated individuals (vs. vaccinated individuals) and individuals experiencing initial infection (vs. individuals experiencing a second or third infection) had an increased risk of infection with SARS-CoV-2.Conclusion: This study analyzed the household SARs and risk factors. We hope that the results can help develop age-specific immunization plans and responses to reduce the SAR in preparation for emerging infectious diseases or potential new variants of SARS-CoV-2.

Developing Risk Factor-Related Injury Indicators Based on Injury Surveillance Data in Korea

이혜아,이선화,임도희,박보현,한혜진,손혜리,안선희,최용주,김보애,임도상,홍성옥,김영택,박주옥,박혜숙 사단법인 대한보건협회 2017 대한보건연구 Vol.43 No.4

Objective : In Korea, the Emergency Department-based In-depth Injury Surveillance System was designed to provide information necessary to establish injury prevention and management policies. To optimize the effectiveness and strategic utility of the injury surveillance system requires risk factor-related injury indicators. Thus, this study aimed to suggest indicators with description of a process to develop indicators that adhere to the goals of a surveillance system using data from hospital-based injury surveillance systems. Methods : Stepwise methodology using a Haddon matrix-based approach was used to develop risk factor- related injury indicators. Expert evaluation of the appropriateness, importance, clarity, sensitivity and applicability of candidate indicators was conducted in two rounds by referencing results for descriptive analysis using injury surveillance data. Results : Through the expert evaluation, 33 indicators were selected including those that measured the scale of injury. When classified according to the injury categories, indicators involving transport accidents were the most common with nine indicators, followed by those related to fall injuries with six indicators. Indicators also included poisoning, drowning, self-injury/suicide and violence. As a risk factor-related indicator, distribution of poisonous substances among injury patients admitted to an emergency department for intentional poisoning was selected. Conclusions : Appropriate indicators are needed that allow the effective operation of an injury surveillance system and policy-based use of such a system. It is also necessary to improve indicators that are cyclic in nature and can reflect changes in social interests and policies.

지역사회건강조사의 조사 기획과 수행

김영택,최보율,이계오,김호,전진호,김수영,이덕형,김윤아,임도상,강영화,이태영,김정숙,조현,김유진,고윤실,서순려,박노례,이종구 대한의사협회 2012 대한의사협회지 Vol.55 No.1

In 2008, the Korean Centers for Disease Control and Prevention (KCDC) initiated Korean Community Health Survey (KCHS), the first nationwide survey to provide data that could be used to plan, implement, monitor and evaluate community health promotion and disease prevention program. This community-based cross-sectional survey has been conducted by 253 community health centers, 36 community universities and 1,500 interviewers. The KCHS standardized questionnaire is developed jointly by KCDC staff, a working group of health indicators standardization subcommittee and 16 metropolitan cities and provinces with 253 regional sites. The KCHS was administered by trained interviewers and the quality control of KCHS was improved by introduction of computer-assisted personal interview in 2010. The questionnaire was reviewed annually so that revised and/or new questions could be added based on public health policy. The additional questions included the fixed and rotating cores, emerging issues and optional modules. The standardized questionnaire of KCHS covered a wide variety of health topics, which could be used to assess the prevalence of personal health behaviors related to causes of disease. The KCHS data allows that the differences of health issues among provinces can be directly compared. Furthermore, the provinces can use these data for their own cost-effective health interventions to improve health promotion and disease prevention.

2008년 지역사회건강조사 질관리정도 평가

김종희 ( Jong Hee Kim ),박해용 ( Hye Yong Park ),김윤아 ( Yun A Ki ),강양화 ( Yang Hwa Gang ),김정숙 ( Jung Suk Kim ),임도상 ( Do Sang Im ),탁양주 ( Yang Ju Tak ),임현술 ( Hyun Sul Im ),김호 ( Ho Kim ) 한국보건정보통계학회(구 한국보건통계학회) 2010 한국보건정보통계학회지 Vol.35 No.2

Objective: 2008 Community health survey is an survey to produce health statistics of 253 communities. This study aims to develope quality management`s system and assess the quality of [2008 Community Health Survey]. Method: We reviewed literatures related to health interview survey in and out countries and visited the survey site to get information. We set up 3-step system for managing quality of [2008 Community health survey], which consists of community health center, college and KCDC. We also developed a survey manuals, a quality management system and a quality index. Using quality-evaluation index, we assessed quality of [2008 Community Health Survey]. Result: Out of 231,355 participants which were 101,307 male, 118,941 female, 220,248(95.2%) completed the questionnaire. Spot-replacement, Household-replacement was 0.6%, 7.2%, respectively. Kappa index of answer-concordance was 0.68 which means a good concordance. According to results of survey site evaluation, quality poor items were Questioning order(Insufficiency 45.6%), Question completion(Insufficiency 36.8%), Explanation of survey goal(Insufficiency 31.6%), Probing(Insufficiency 28.1%) orderly. Conclusion: The quality of [2008 Community Health Survey] was fine, but more efforts to improve health-survey quality need in the future.