• 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.

• 응용통계연구
• /
• 제36권1호
• /
• pp.1-19
• /
• 2023

본 연구는 학생 간의 연결에 어떤 속성이 유효한지 종단 네트워크 분석을 통해 알아보고자 하였으며, 종단 네트워크 모형인 TERGM (temporal exponential random graph model)과 SAOM (stochastic actor-oriented model) 통계적 모형을 사용하고 결과를 비교하였다. TERGM 모형은 네트워크 전체의 연결 형성을 바탕으로, SAOM 모형은 특정 행위자가 형성하는 주변 네트워크를 대상으로 연구 결과를 해석하였다. TERGM 모형은 시간 항을 통해 이전 시점의 영향을 표현하였으며, SAOM 모형은 비율 함수로 행위자의 기회에 의해 진화하는 네트워크를 구현해 시간적 종속성을 고려하였다.

• 기술경영경제학회:학술대회논문집
• /
• 기술경영경제학회 2003년도 제22회 동계학술발표회 논문집
• /
• pp.63-77
• /
• 2003

In this paper, we investigate the co-movements of the real GDP cycle and the telecommunications service expenditure cycle over 1980 through 2000 in Korea. We used several appropriate statistical tools to characterize and to compare the co-movements between the two aforementioned variables. The statistics presented in this paper are common, easy to Interpret and capture important visual, descriptive information and the dynamics of data that would be lost if one focused only on the unconditional correlation coefficients of de-trended data. This paper examines four possible characteristics for the co-movements of real GDP and telecommunications service expenditure in Korea. The first is that in low frequency, the cyclical movements in telecommunications service expenditure are very similar to the cyclical movements in real GDP, The second is that in the high frequency, the estimated stationary cyclical telecommunications service expenditure is not closely related to the estimated stationary cyclical real GDP. Third, according to co-spectra analysis, the strong relationships between the variables are found in the mid-term frequency and as such this empirical finding has important implications for telecommunications policy and telecommunications service implementation strategies. Finally, in the short-term period, our empirical results support telecommunications network theories such as path dependence properties, lock-in, positive feedback externality, and network externality.

• 응용통계연구
• /
• 제27권3호
• /
• pp.475-485
• /
• 2014

일반화 선형모형(GLM)에 기초한 확률적 날씨 발생기(Stochastic weather generator)는 일일 날씨를 생성하는데 가장 일반적으로 사용되는 방법인다. 본 논문에서는 다층구조를 이용하여 기존의 GLM weather generator에 공간구조를 소개하였다. 계절별 총강우량의 overdispersion 현상을 효과적으로 제거하기 위해서 smoothing된 계절별 총강우량을 모형에 포함하였고 공간구조를 소개하기 위해서 Stochastic weather generator의 모형계수에 공간구조를 가지는 다변량 정규분포를 가정하였다. 그리고 제안된 공간구조를 가지는 GLM weather generator 모형을 우리나라 76개 지역에서 39년간 측정된 일별 강우량 관측자료에 적용하였다.

• 한국치위생학회지
• /
• 제15권5호
• /
• pp.889-898
• /
• 2015

Objectives: The purpose of the study is to investigate the oral health-related quality of life (OHIP-14) according to smoking in the male workers in Gunsan. Methods: A self-reported questionnaire was completed by 460 male workers in Gunsan from April 13 to 30, 2015. Among 460 workers, smokers were 205 and nonsmokers were 255. The questionnaire included three questions of general characteristics of the subjects, four questions of subjective oral health, fourteen questions of OHIP, ten questions of sobriety test (AUDIT), and eight questions of smoking (Fagerstrom Test for Nicotine Dependence, FTND). Data were analyzed by frequency analysis, t-test, ANOVA, Pearson's correlation, and multiple regression analysis using IBM SPSS(Statistical Package for the Social Science) for Windows 20.0 program. Results: In the nonsmokers, there was a negative correlation between the alcohol consumption, dental health condition, tooth pain, gum bleeding, and halitosis. The higher OHIP in the smokers had the negative correlation with nicotine dependence, alcohol consumption, tooth pain, gum bleeding, and halitosis. In order to improve OHIP, smokers are encouraged to manage gum bleeding, and non-smokers to halitosis. Conclusions: In order to improve OHIP, cessation of smoking and alcohol consumption is very important and it can lead to improve the quality of life in the workers.

• 대한가정학회지
• /
• 제20권4호
• /
• pp.33-44
• /
• 1982

The purpose of this study was to investigate clothing behavior of women in Dae Jeon and to find out the relationship between clothing behavior and occupations. Measures selected for this study was‘Creekmore Scales of Eight Clothing Variables’, (Creekmore, 1966). The questionnaires were distributed to women in Dae Jeon including teachers, women working in a government office, women working in a company, nurses, women working in a bank, women working in a department store and the business center, technicians, women working in a merry making place, and house wives. For statistical analyses of the data from 400 respondents, F-test, and t-test were used. The results were as follows: 1. There were partially significant relationship between clothing behavior and occupations. The significant variables were modesty, comfort, approval, and dependence among the eight clothing variables. a) The group of highest interest in modesty and comfort was house wives and the group of lowest interest was women working in a merry making place. b) The group of highest interest in approval was women working in a merry making place and the group of lowest interest was women working in a bank. c) The group of highest interest in dependence was teachers and the group of lowest interest was nurses. 2. Partially significant differences were found in clothing behavior between a working women and house wives: The house wives were more interest in modesty, comfort, management. 3. Partially significant difference were found in clothing behavior between White-collar workers and Blue-collar workers were more interest in attention, approval, interest.

• 한국통신학회논문지
• /
• 제28권7B호
• /
• pp.620-627
• /
• 2003

EPON망은 중앙 기지국과 가입자들 사이에 광 액세스 라인을 저렴하게 제공할 수 있는 액세스 망 기술로 떠오르고 있다. 동적 대역 할당은 상향 채널의 효율적인 활용을 위해 ONU간 통계적 다중화를 제공한다. 본 논문에서는 EPON망에서 차등 CoS를 제공하기 위해 주기적 폴링을 기반으로 한 동적 대역 할당 방법을 제안한다. 더불어, Interleaved 폴링 방식에 기반한 동적 대역 할당 방안은 상향 트래픽 로드가 적은 경우 하향 채널의 용량을 대폭 감소시키는 단점을 가지고 있음을 보인다. 실제적인 시뮬레이션 결과를 얻기 위해 self-similarity와 long-range dependence 특성을 갖는 트래픽을 이용하였다. EPON망의 성능은 ONU에 제공되는 다양한 로드별로 분석하였으며, 특히 차등 CoS 특성 분석을 위해 등급별 패킷의 지연 특성을 분석하였다.

• 아태비즈니스연구
• /
• 제13권3호
• /
• pp.171-182
• /
• 2022

Purpose - The main purpose of this study is to test two competing hypotheses on the relationship between foreignness and corporate philanthropy drawing on the resource dependence theory. Design/methodology/approach - This study used Korean Enterprsie Survey data constructed by Korean Statistical Information Service. This study employed both OLS and Tobit models to test the hypotheses. Findings - A conventional wisdom can argue that MNC subsidiaries are more willing to give charitable donations than local firms do because they may perceive philanthropy in the host country is a good opportunity to reduce liabilities of foreignness (H1-1). On the other hand, one can argue that MNC subsidiaries are less willing to give charitable donations than local firms do because they have operational flexibility when they are coerced to engage in corporate philanthropy by host country government stakeholders (H1-2). Empirical results support the prediction that MNC subsidiaries are less likely to give charitable donations than local firms do in Korea. Research implications or Originality - Our findings can provide useful insights to researchers and managers in the sense that MNC managers can increase or decrease their charitable donations depending on the institutional contingencies in different host countries. These institutional contingencies are particularly important within a transforming context such as Korea.

• 한국지역사회생활과학회지
• /
• 제27권4호
• /
• pp.803-815
• /
• 2016

The purpose of this study is to increase our understanding of subjective well-being in adolescents and to help improve it by investigating the relationship among adolescents' temperaments, perceived parenting affective attitude and subjective well-being. For this study, the subjects were 354 students from 4 different middle schools in Gwang-ju, Korea. The statistical analysis of the data was conducted by means of the SPSS 21.0 program using Pearson's correlations and hierarchical regression. The results are as follows: First, the temperament and perceived parenting affective attitude were correlated with subjective well-being. Among the temperament traits, novelty seeking and harm avoidance showed negative correlations with subjective well-being, whereas affection, reward dependence and persistence showed positive correlations with subjective well-being. Second, after the mediating effect of perceived affection on subjective well-being was confirmed, three temperament traits, viz. harm avoidance, reward dependence, and persistence, were identified as mediating factors. The conclusion of this study is that adolescents' temperament and perceived parenting affective attitude are related to their subjective well-being and can also predict their subjective well-being. Harm avoidance, reward dependence, and persistence, which are inborn temperaments, affect subjective well-being. Affection, which is a perceived parenting affective attitude trait, affects subjective well-being as a mediation variable. This study also suggests that it is more effective to influence perceived parenting affective attitude, which has growth potential, rather than temperaments which are hard to change.

• Communications for Statistical Applications and Methods
• /
• 제30권2호
• /
• pp.119-133
• /
• 2023

With the rapid growth of the economy and fossil fuel consumption, the concentration of air pollutants has increased significantly and the air pollution problem is no longer limited to small areas. We conduct statistical analysis with the actual data related to air quality that covers the entire of South Korea using R and Python. Some factors such as SO₂, CO, O₃, NO₂, PM₁₀, precipitation, wind speed, wind direction, vapor pressure, local pressure, sea level pressure, temperature, humidity, and others are used as covariates. The main goal of this paper is to predict air quality index (AQI) spatio-temporal data. The observations of spatio-temporal big datasets like AQI data are correlated both spatially and temporally, and computation of the prediction or forecasting with dependence structure is often infeasible. As such, the likelihood function based on the spatio-temporal model may be complicated and some special modelings are useful for statistically reliable predictions. In this paper, we propose several methods for this big spatio-temporal AQI data. First, random effects with spatio-temporal basis functions model, a classical statistical analysis, is proposed. Next, neural networks model, a deep learning method based on artificial neural networks, is applied. Finally, random forest model, a machine learning method that is closer to computational science, will be introduced. Then we compare the forecasting performance of each other in terms of predictive diagnostics. As a result of the analysis, all three methods predicted the normal level of PM_2.5 well, but the performance seems to be poor at the extreme value.