Bayesian smoothing under structural measurement error model with multiple covariates

Jinseub Hwang,Dal Ho Kim 한국데이터정보과학회 2017 한국데이터정보과학회지 Vol.28 No.3

In healthcare and medical research, many important variables have a measurement error such as body mass index and laboratory data. It is also not easy to collect samples of large size because of high cost and long time required to collect the target patient satisfied with inclusion and exclusion criteria. Beside, the demand for solving a complex scientific problem has highly increased so that a semiparametric regression approach could be of substantial value solving this problem. To address the issues of measurement error, small domain and a scientific complexity, we conduct a multivariable Bayesian smoothing under structural measurement error covariate in this article. Specifically we enhance our previous model by incorporating other useful auxiliary covariates free of measurement error. For the regression spline, we use a radial basis functions with fixed knots for the measurement error covariate. We organize a fully Bayesian approach to fit the model and estimate parameters using Markov chain Monte Carlo. Simulation results represent that the method performs well. We illustrate the results using a national survey data for application.

Bayesian Curve-Fitting in Semiparametric Small Area Models with Measurement Errors

Hwang, Jinseub,Kim, Dal Ho The Korean Statistical Society 2015 Communications for statistical applications and me Vol.22 No.4

We study a semiparametric Bayesian approach to small area estimation under a nested error linear regression model with area level covariate subject to measurement error. Consideration is given to radial basis functions for the regression spline and knots on a grid of equally spaced sample quantiles of covariate with measurement errors in the nested error linear regression model setup. We conduct a hierarchical Bayesian structural measurement error model for small areas and prove the propriety of the joint posterior based on a given hierarchical Bayesian framework since some priors are defined non-informative improper priors that uses Markov Chain Monte Carlo methods to fit it. Our methodology is illustrated using numerical examples to compare possible models based on model adequacy criteria; in addition, analysis is conducted based on real data.

Multivariable Bayesian curve-fitting under functional measurement error model

Jinseub Hwang,Dal Ho Kim 한국데이터정보과학회 2016 한국데이터정보과학회지 Vol.27 No.6

A lot of data, particularly in the medical field, contain variables that have a mea-surement error such as blood pressure and body mass index. On the other hand, recently smoothing methods are often used to solve a complex scientific problem. In this paper, we study a Bayesian curve-fitting under functional measurement error model. Espe-cially, we extend our previous model by incorporating covariates free of measurement error. In this paper, we consider penalized splines for non-linear pattern. We employ a hierarchical Bayesian framework based on Markov Chain Monte Carlo methodology for fitting the model and estimating parameters. For application we use the data from the fifth wave (2012) of the Korea National Health and Nutrition Examination Survey data, a national population-based data. To examine the convergence of MCMC sam-pling, potential scale reduction factors are used and we also confirm a model selection criteria to check the performance.

Multivariable Bayesian curve-fitting under functional measurement error model

Hwang, Jinseub,Kim, Dal Ho The Korean Data and Information Science Society 2016 한국데이터정보과학회지 Vol.27 No.6

A lot of data, particularly in the medical field, contain variables that have a measurement error such as blood pressure and body mass index. On the other hand, recently smoothing methods are often used to solve a complex scientific problem. In this paper, we study a Bayesian curve-fitting under functional measurement error model. Especially, we extend our previous model by incorporating covariates free of measurement error. In this paper, we consider penalized splines for non-linear pattern. We employ a hierarchical Bayesian framework based on Markov Chain Monte Carlo methodology for fitting the model and estimating parameters. For application we use the data from the fifth wave (2012) of the Korea National Health and Nutrition Examination Survey data, a national population-based data. To examine the convergence of MCMC sampling, potential scale reduction factors are used and we also confirm a model selection criteria to check the performance.

Bayesian smoothing under structural measurement error model with multiple covariates

Hwang, Jinseub,Kim, Dal Ho The Korean Data and Information Science Society 2017 한국데이터정보과학회지 Vol.28 No.3

In healthcare and medical research, many important variables have a measurement error such as body mass index and laboratory data. It is also not easy to collect samples of large size because of high cost and long time required to collect the target patient satisfied with inclusion and exclusion criteria. Beside, the demand for solving a complex scientific problem has highly increased so that a semiparametric regression approach could be of substantial value solving this problem. To address the issues of measurement error, small domain and a scientific complexity, we conduct a multivariable Bayesian smoothing under structural measurement error covariate in this article. Specifically we enhance our previous model by incorporating other useful auxiliary covariates free of measurement error. For the regression spline, we use a radial basis functions with fixed knots for the measurement error covariate. We organize a fully Bayesian approach to fit the model and estimate parameters using Markov chain Monte Carlo. Simulation results represent that the method performs well. We illustrate the results using a national survey data for application.

Bayesian curve-fitting with radial basis functions under functional measurement error model

Hwang, Jinseub,Kim, Dal Ho The Korean Data and Information Science Society 2015 한국데이터정보과학회지 Vol.26 No.3

This article presents Bayesian approach to regression splines with knots on a grid of equally spaced sample quantiles of the independent variables under functional measurement error model.We consider small area model by using penalized splines of non-linear pattern. Specifically, in a basis functions of the regression spline, we use radial basis functions. To fit the model and estimate parameters we suggest a hierarchical Bayesian framework using Markov Chain Monte Carlo methodology. Furthermore, we illustrate the method in an application data. We check the convergence by a potential scale reduction factor and we use the posterior predictive p-value and the mean logarithmic conditional predictive ordinate to compar models.

Statistical analysis of KNHANES data with measurement error models

Hwang, Jinseub The Korean Data and Information Science Society 2015 한국데이터정보과학회지 Vol.26 No.3

We study a statistical analysis about the fifth wave data of the Korea National Health and Nutrition Examination Survey based on linear regression models with measurement errors. The data is obtained from a national population-based complex survey. To demonstrate the availability of measurement error models, two results between the general linear regression model and measurement error model are compared based on the model selection criteria which are Akaike information criterion and Bayesian information criterion. For our study, we use the simulation extrapolation algorithm for measurement error model and the jackknife method for the estimation of standard errors.

구조적 측정오차를 고려한 준모수적 Fay-Herriot 모형

류수락(Soorack Ryu),황진섭(Jinseub Hwang) 한국자료분석학회 2018 Journal of the Korean Data Analysis Society Vol.20 No.1

대부분의 연구에서는 측정을 통하여 변수를 생성하게 되며, 이러한 변수들 중에는 측정오차를 가지고 있는 경우가 발생할 수 있다. 이러한 측정오차는 통계적 분석을 복잡하게 하며 이러한 문제를 일반적으로 측정오차 문제라고 한다. 또한 현재의 과학적인 현상들은 단순한 선형적 관계가 아닌 복잡한 관련성을 증명하고자 하고 있으며, 이를 해결하기 위한 여러 가지 비선형적 모형들이 개발되고 있다. 이에 사전 연구(Ryu, Hwang, 2017)에서는 측정오차를 가지는 공변량의 참값(true value)에 대한 비확률성(non-stochastic)을 가정하는 기능적 측정오차모형(functional measurement error model)과 비선형성을 고려한 준모수적 Fay-Herriot 모형을 제안하였다. 본 연구에서는 이를 확장하여 측정오차를 가지는 공변량의 참값에 확률성(stochastic)을 가정하는 구조적 측정오차모형(structural measurement error model)을 고려한 준모수적 Fay-Herriot 모형을 제안하고자 한다. 모형 적합 및 모수 추정에서는 MCMC(Markov chain Monte Carlo) 방법을 사용하는 계층적 베이지안 모형을 고려하였다. 두 가지의 비선형적 함수를 이용한 모의실험을 통하여 본 연구에서 제안한 모형의 우수성을 확인하였으며, 실증자료의 적용을 위해 국민건강영양조사의 제7기 1차연도(2016) 자료를 사용하였다. In most studies, variables are created through measurement and some of these variables may have measurement errors. This measurements error complicates the statistical analysis and this problem is commonly called measurement error problem. Also, current scientific phenomena are trying to demonstrate the complexity of relationships, not a simple linear relationships and then various nonlinear models are being developed. In our previous study (Ryu, Hwang, 2017) we proposed a semiparametric Fay-Herriot model under the functional measurement error that assumes a non-stochastic true value of the covariate. In this study, we propose a semiparametric Fay-Herriot model under structural measurement error that assumes a stochastic true value of the covariate as an extension of our previous model. For the model fitting and parameter estimation, we consider hierarchical Bayesian model approach using Markov chain Monte Carlo method. To check the superiority of the proposed model, we conduct simulation studies using two nonlinear functions and we use the seventh KNHANES (Korean national health and nutrition examination survey) data for the application.

Urban-Rural Differences in Prevalence of Depressive Symptoms and Its Related Factors Among Older Adults: Findings from the Korean Longitudinal Study of Aging

김봉정,Hwang Jinseub,김도향,강수진 한국지역사회간호학회 2024 지역사회간호학회지 Vol.35 No.1

Purpose: Many studies exist on factors associated with depressive symptoms in urban and rural older adults; however, studies using a longitudinal design are scarce. This study aimed to determine whether there is a difference in the prevalence of depressive symptoms and their associated risk factors over time between urban and rural areas using a longitudinal sample of Korean older adults. Methods: Data from the Korean Longitudinal Study of Aging (2006–2020) of older adult participants ≥65 years without depressive symptoms were analyzed. A generalized estimating equation model was employed for repeated measures analysis. Results: As the time of living in the area increased, the prevalence risk of depressive symptoms in older adults increased in urban areas compared to rural areas in adjusted Model 2. In urban areas, less social contact with neighbors was significantly associated with a higher risk of depressive symptoms. In rural areas, occasional social contact with children was significantly associated with a lower risk of depressive symptoms. The prevalence of depressive symptoms was associated with five components of successful aging with some variations observed based on urban and rural residential areas. Conclusions: Our findings suggest that understanding the longitudinal impact of residence on depressive symptoms provides valuable insights into the relationship between urban/rural areas and depressive symptoms. This study highlights the need for nursing intervention efforts aimed at promoting successful aging and increasing social contact with children or neighbors.

성향점수 분석방법에 따른 대구광역시 학생들의 교과목별 사교육 효과

류수락(Soorack Ryu),황진섭(Jinseub Hwang) 한국데이터정보과학회 2019 한국데이터정보과학회지 Vol.30 No.4

대구시 학생들의 사교육 효과에 대한 많은 기존 연구들이 있지만 대부분의 연구들은 표본설계를 통해 수집된 대표성 있는 자료가 아니고 결과들을 일반화할 수 없는 한계점을 가지고 있다. 이에 본 연구에서는 대구광역시교육연구정보원에서 표본설계를 통해 학생, 학부모, 교사를 대상으로 설문을 진행한 2016년 대구교육실태조사 자료를 기반으로 대상자들의 교란요인을 보정하고자 성향점수의 여러가지 분석방법(매칭, 공변량보정, 역확률가중치)을 활용하여 초중고 학생들의 교과목별 (국어, 수학, 영어) 사교육 참여 여부에 따른 학업성취도와 수업이해도, 전국연합학력평가 효과를 확인하고자 한다. 최종 연구대상자는 초등학생 410명, 중학생 571명, 고등학생 913명이었으며 연구 결과를 살펴보면, 초등학생에서는 역확률 가중치 방법을 적용하는 경우 수학에서 학업성취도와 수업이해도에 대한 유의한 사교육효과가 나타났다. 중학생에서는 매칭 방법을 적용한 경우 수학과 영어의 수업이해도, 역확률 가중치 방법을 적용한 경우 국어의 수업이해도를 제외하고 모두 유의한 효과가 나타났다. 고등학생에서는 매칭 방법을 적용한 경우 영어의 학업성취도를 제외하고 모두 유의한 효과가 나타났으며, 역확률 가중치 방법을 적용한 경우 국어와 수학의 학업성취도와 전국연합학력평가, 공변량 보정방법을 적용한 경우 수학의 학업성취도에 유의한 효과가 나타났다. 이와 같이 성향점수의 분석방법에 따라 학력별 및 교과목별 사교육의 효과는 일부 다르게 나타나고 있으므로 분석방법의 특징을 고려하여 연구목적에 적합한 성향점수 방법의 선택이 필요하다. In this study we use the 2016 educational survey data of Daegu Educational Research and Information Institute and we confirm the effect of private tutoring for elementary, middle and high school students in Daegu about the academic achievement, learning understanding and national mock test based on propensity score (matching, covariate adjusting and IPTW (inverse probability of treatment weighting)). According to the results, in elementary school, there are significant effects for the academic achievement and learning understanding of Math when using IPTW. In middle school, it is an effective for the learning understanding of Math and English when using matching, and in all except for the learning understanding of Korean based on IPTW. In high school, all except the academic achievement of English have significant effect when using matching and there are significant effects for the academic achievement and national mock test in Korean and Math, and the academic achievement in Math based on IPTW and covariate adjusting, respectively. The effectiveness of a private tutoring by grade and subject is different according to the method, so it is necessary to select the propensity score method appropriate for the research purpose considering the characteristics of methods.