• Communications for Statistical Applications and Methods
• /
• 제11권2호
• /
• pp.381-397
• /
• 2004

The marginal likelihood has become an important tool for model selection in Bayesian analysis because it can be used to rank the models. We discuss the marginal likelihood for Poisson regression models that are potentially useful in small area estimation. Computation in these models is intensive and it requires an implementation of Markov chain Monte Carlo (MCMC) methods. Using importance sampling and multivariate density estimation, we demonstrate a computation of the marginal likelihood through an output analysis from an MCMC sampler.

• Journal of the Korean Statistical Society
• /
• 제26권4호
• /
• pp.493-505
• /
• 1997

The analysis of binary data appears to many areas such as statistics, biometrics and econometrics. In many cases, data are often collected in which some observations are incomplete. Assume that the missing covariates are missing at random and the responses are completely observed. A method to Bayesian analysis of the binary regression model with incomplete data is presented. In particular, the desired marginal posterior moments of regression parameter are obtained using Meterpolis algorithm (Metropolis et al. 1953) within Gibbs sampler (Gelfand and Smith, 1990). Also, we compare logit model with probit model using Bayes factor which is approximated by importance sampling method. One example is presented.

• 한국콘텐츠학회논문지
• /
• 제13권4호
• /
• pp.53-63
• /
• 2013

본 논문은 효율적인 몬테 칼로 시뮬레이션을 위하여 중요 샘플링(Importance sampling) 기법이 내장된 실험 틀을 제안한다. 제안하는 실험 틀은 중요 샘플링 기법을 적용하기 위해 기능적으로 세분화된 중요 표본기(Importance Sampler)와 편향 보상기(Bias Compensator)라는 두 개의 하위 모델을 내장(Embedded)한다. 이러한 하위 모델은 기존의 시스템 모델과 실험 틀의 경계에 플러그인 됨으로써 기존 모델들의 수정없이 재사용할 수 있는 장점이 있다. 그리고 제안하는 실험 틀은 기능적 측면에서 중요 사건에 대하여 동일한 수준의 결과를 얻는 데 있어 기존의 몬테 칼로 시뮬레이션보다 시뮬레이션 시간을 단축시킬 수 있다. 이러한 효용성을 입증하기 위해 두 가지 실험을 수행하였는데, 실험 결과, 본 실험에 대하여 기존의 몬테칼로 시뮬레이션보다 최대 400 배 가량의 시뮬레이션 시간 측면에서 성능 향상을 확인하였다. 본 논문에서 제안하는 실험 틀은 다양한 콘텐츠 분야에 적용되어 시뮬레이션 성능을 향상시킬 수 있는 도구로 활용할 수 있을 뿐 아니라, 교육적 측면에서 다양한 사회 현상을 이해하고 해석하는 도구로도 활용이 가능하다.

• Communications for Statistical Applications and Methods
• /
• 제29권1호
• /
• pp.65-83
• /
• 2022

Although various statistical methods have been developed to map time-dependent genetic factors, most identified genetic variants can explain only a small portion of the estimated genetic variation in longitudinal traits. Gene-gene and gene-time/environment interactions are known to be important putative sources of the missing heritability. However, mapping epistatic gene-gene interactions is extremely difficult due to the very large parameter spaces for models containing such interactions. In this paper, we develop a Gaussian process (GP) based nonparametric Bayesian variable selection method for longitudinal data. It maps multiple genetic markers without restricting to pairwise interactions. Rather than modeling each main and interaction term explicitly, the GP model measures the importance of each marker, regardless of whether it is mostly due to a main effect or some interaction effect(s), via an unspecified function. To improve the flexibility of the GP model, we propose a novel grid-based method for the within-subject dependence structure. The proposed method can accurately approximate complex covariance structures. The dimension of the covariance matrix depends only on the number of fixed grid points although each subject may have different numbers of measurements at different time points. The deviance information criterion (DIC) and the Bayesian predictive information criterion (BPIC) are proposed for selecting an optimal number of grid points. To efficiently draw posterior samples, we combine a hybrid Monte Carlo method with a partially collapsed Gibbs (PCG) sampler. We apply the proposed GP model to a mouse dataset on age-related body weight.

• Journal of the Korean Data and Information Science Society
• /
• 제15권1호
• /
• pp.227-237
• /
• 2004

In this paper we presented a Bayesian inference approach for estimating the location and scale parameters of the lognormal distribution using iterative Gibbs sampling algorithm. We also presented estimation of location parameter by two non iterative methods, importance sampling and weighted bootstrap assuming scale parameter as known. The estimates by non iterative techniques do not depend on the specification of hyper parameters which is optimal from the Bayesian point of view. The estimates obtained by more sophisticated Gibbs sampler vary slightly with the choices of hyper parameters. The objective of this paper is to illustrate these tools in a simpler setup which may be essential in more complicated situations.

• 한국산업보건학회지
• /
• 제34권2호
• /
• pp.115-124
• /
• 2024

Objectives: The aim of this study is to assess indoor air quality within and around buildings and evaluate the health risks associated with exposure to indoor air pollution. The study compares IAQ standards established by the World Health Organization with those set by South Korea's Ministry of Environment and Ministry of Education. Methods: The study utilized an Anderson Sampler and DustTrak^TM II to collect samples of total airborne bacteria and PM in indoor and outdoor environments. Collected samples were analyzed using biological and biochemical methods. Statistical analysis was conducted using SPSS to examine the correlation between airborne bacteria and PM. Results: The study revealed that the concentration of total airborne bacteria in indoor air generally remained below the Ministry of Environment's standard of 800 CFU/m³, although it surpassed this threshold in certain instances. PM concentrations did not exceed the standards. Indoor fine dust concentration was higher when there were people (P<0.05). There was no difference in total floating bacterial concentrations between indoor and outdoor environments (P=0.184). Finally, there was a correlation between fine dust and airborne bacteria concentrations. Conclusion: The study evaluated the concentrations of total airborne bacteria and PM in indoor air, emphasizing the importance of managing IAQ. Further research in various environments is essential to ensure a healthy indoor environment. The findings underscore the need for ongoing research and management to enhance IAQ and create safer and healthier living environments.

• Journal of the Korean Statistical Society
• /
• 제31권2호
• /
• pp.177-198
• /
• 2002

The estimation of variance components or variance ratios in linear model is an important issue in plant or animal breeding fields, and various estimation methods have been devised to estimate variance components or variance ratios. However, many traits of economic importance in those fields are observed as dichotomous or polychotomous outcomes. The usual estimation methods might not be appropriate for these cases. Recently threshold linear model is considered as an important tool to analyze discrete traits specially in animal breeding field. In this note, we consider a hierarchical Bayesian method for the threshold animal model. Gibbs sampler for making full Bayesian inferences about random effects as well as fixed effects is described to analyze jointly discrete traits and continuous traits. Numerical example of the model with two discrete ordered categorical traits, calving ease of calves from born by heifer and calving ease of calf from born by cow, and one normally distributed trait, birth weight, is provided.

• 한국수산과학회지
• /
• 제22권4호
• /
• pp.189-195
• /
• 1989

본 연구는 1982년 8월부터 1986년 2월 까지 5년에 걸쳐 계절별로 한반도에 인접한 황해의 저서다모류를 대상으로 정량 채집하여 다모류의 중요성과 종조성을 파악하였다. 다모류의 비중은 전 저서동물군 중에서 가장 높은 $35.1\~46.4\%$의 조성률을 나타내었고, 생체양에 있어서는 극피동물이나 연체동물군이 우점군으로 나타났으나 그들의 계절적 변화 폭이 일정하지 않고 매우 다양했다. 채집도구에 따른 다모류의 비중은 드레지(dre-dge)를 사용할 경우 채이기(grab)의 경우보다 약간 낮은 조성률을 나타내었다. 이는 드레지의 경우가 훨씬 더 퇴적물의 표층부분만을 예인하기 때문이었으며 다모류의 조성률도 다소 낮은 경향을 보였다. 본 연구기간 중 출현한 다모류는 모두 141종으로 나타났으며 이 해역에서의 대표종은Spiophanes bormbyx, Ampharete arctica, Geniada maulata, Nephtys caeca, Nothria iridescens이었다.

• 한국대기환경학회지
• /
• 제15권4호
• /
• pp.485-493
• /
• 1999

This paper presents the results of the first interlaboratory comparison of ambient particles measurement conducted by 5 laboratories under the framework of Korean Society for Atmospheric Environment(KOSAE) on the May of 1998 in Seoul. Collection of particles less than 10$mu extrm{m}$(PM10) was carried out by two different types of PM10 sampler, and their concentration reported by three laboratories ranged from 36.2~56.05$\mu\textrm{g}$/㎥. Analyses of chemical composition of particles collected by each lab were performed by the destructive and nondestructive instrumental methods. Statistical analysis of the measurement results could not be carried out because a comprehensive QA/QC had not been applied before experiment. The importance of QA/QC as well as the performance of QA/QC for the measurement of PM10 was discussed in detail.

• 대한예방의학회:학술대회논문집
• /
• 대한예방의학회 1994년도 교수 연수회(환경)
• /
• pp.585-595
• /
• 1994

The region of the respiratory tract where inhaled particles deposit can have important implications for the causation of local or systemic toxic effects. For most aerosols of occupational importance, respiratory tract deposition can be predicted from the aerodynamic diameter of the particles. With the advent of cascade impactors, particularly those of personal sampler size, the determination of the aerodynamic diameters of aerosols has become more common. Some limitations of cascade impactor use are well recognized (e.g., particle bounce and substrate overloading) and are generally correctable. However, two important limitations of the instruments may not be receiving adequate attention: relative humidity effects on potentially hygroscopic aerosols and the collection characteristics of fibrous aerosols as compared to their actual deposition site potential. The results of this study, when compared to results of previous controlled laboratory trials, suggest that, while potentially hygroscopic lead aerosols from lead acid battery plant operations do not appear to be affected by changes in plant environmental humidity levels, the potential - exists for significant size changes upon inhalation. Secondly, fibers were detected in aerodynamic size ranges that would be associated with deep lung deposition; however, upon microscopic examination, these same fibers would actually be predicted to deposit in the upper airways. This study suggests that the physicalchemical properties and morphological features of an aerosol should be carefully considered by industrial hygienists before cascade impactors are used in attempts to predict the effects of inhaled aerosols.