• Journal of the Korean Data and Information Science Society
• /
• 제27권3호
• /
• pp.815-825
• /
• 2016

Marginalized random effects models (MREMs) are often used to analyze longitudinal categorical data. The models permit direct estimation of marginal mean parameters and specify the serial correlation of longitudinal categorical data via the random effects. However, it is not easy to estimate the random effects covariance matrix in the MREMs because the matrix is high-dimensional and must be positive-definite. To solve these restrictions, we introduce two modeling approaches of the random effects covariance matrix: partial autocorrelation and the modified Cholesky decomposition. These proposed methods are illustrated with the real data from Korean genomic epidemiology study.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2006년도 Proceedings of ISRS 2006 PORSEC Volume II
• /
• pp.892-895
• /
• 2006

The atmospheric effects on the retrieval of sea ice concentration from passive microwave sensors are examined using simulated data typical for the Arctic summer. The simulation includes atmospheric contributions of cloud liquid water and water vapor and surface wind on surface emissivity on the microwave signatures. A plane parallel radiative transfer model is used to compute brightness temperatures at SSM/I frequencies over surfaces that contain open water, first-year (FY) ice and multi-year (MY) ice and their combinations. Synthetic retrievals in this study use the NASA Team (NT) algorithm for the estimation of sea ice concentrations. This study shows that if the satellite sensor’s field of view is filled with only FY ice the retrieval is not much affected by the atmospheric conditions due to the high contrast between emission signals from FY ice surface and the signals from the atmosphere. Pure MY ice concentration is generally underestimated due to the low MY ice surface emissivity that results in the enhancement of emission signals from the atmospheric parameters. Simulation results in marginal ice areas also show that the atmospheric and surface effects tend to degrade the accuracy at low sea ice concentration. FY ice concentration is overestimated and MY ice concentration is underestimated in the presence of atmospheric water and surface wind at low ice concentration. In particular, our results suggest that strong surface wind is more important than atmospheric water in contributing to the retrieval errors of total ice concentrations over marginal ice zones.

• 응용통계연구
• /
• 제15권1호
• /
• pp.129-137
• /
• 2002

본 논문은 개체의 반응에 영향을 미치는 독립변수들중 일부는 고정요인들이고 일부는 확률요인들로 간주되며 반응변수가 다가범주를 갖는 명목형 변수일 때, 다원분류표에서 자료를 분석하기 위한 모형으로 혼합효과모형을 제시하고 모형내 미지모수들을 추정하는 방법을 다루고 있다

• 대한치과의사협회지
• /
• 제55권11호
• /
• pp.766-776
• /
• 2017

This clinical study was conducted to evaluate the clinical effects of a concave neck of external connection type implant fixture designed for platform switching on the peri-implant tissue responses. Two types of implants with different neck designs were implanted in 20 patients. For the experimental group, the bioseal(BS) implant fixtures with 's' shaped concave profile on the neck were used, and non-bioseal(NBS) implant fixtures with a straight profile on the neck were used as control(Total of 40 implants, NBS: n = 19, BS: n=21). During the one-year period after implant placement, implant survival rate, marginal bone resorption, bleeding, plaque, and complications were evaluated. The survival rate of NBS and BS group was 94.74% and 90.48%, respectively. There was no significant difference on marginal bone resorption, bleeding and plaque between the two groups (P>.05). Within the limits of the present study, implants with a concave neck design showed similar clinical results to implants with a straight neck design on the peri-implant tissue responses. Longitudinal clinical studies are necessary to confirm more effective clinical results.

• Nutrition Research and Practice
• /
• 제2권4호
• /
• pp.283-288
• /
• 2008

This study was carried out to determine whether a short-tenn zinc supplementation contributes to beneficial changes in glycemic control among type 2 diabetic patients. Seventy-six diabetic subjects and 72 normal adults participated in this study. Subjects were divided into supplemented and control groups. Forty-four diabetic patients and 34 normal subjects were supplemented with 50 mg zinc daily as zinc gluconate for 4 weeks. Zinc status was assessed from fasting plasma levels and urinary excretion. The effects of zinc supplementation on fasting blood glucose, $HbA_{1c}$, insulin, and C-peptide were measured at the beginning of the study and after 4 weeks of supplementation. The changes in glycemic control indicators were compared between diabetic groups, classified by baseline $HbA_{1c}$ levels, and by diabetic duration. At baseline, the incidence of marginal zinc deficiency in the diabetic group, as determined by plasma zinc level, was approximately twice as high as in the normal adult group. The changes of $HbA_{1c}$ concentration, and fasting blood glucose following supplementation were not statistically significant in diabetic subjects. In normal subjects, a significant decrease of $HbA_{1c}$ occurred only in the zinc supplemented group. No significant changes were observed for serum insulin and C-peptide in diabetic as well as normal subjects. However, when the changes were compared by baseline $HbA_{1c}$ level, we found that diabetic subjects with $HbA_{1c}\;{\geq}\;7.5%$ showed significantly improved levels of $HbA_{1c}$ and fasting glucose after Zn supplementation. While such improvement in fasting blood glucose was significant among diabetics with shorter diabetic duration, significant levels of increase in serum insulin and C-peptide were observed in zinc supplemented subjects with longer diabetic duration. Fasting blood glucose was significantly decreased, whereas serum insulin and C-peptide were increased in diabetics with marginal zinc status. Therefore, we suggest that Zn supplementation for a short-term period may improve glycemic control in diabetic patients with higher $HbA_{1c}$ levels and marginal zinc status.

• 자원ㆍ환경경제연구
• /
• 제31권2호
• /
• pp.141-163
• /
• 2022

이 논문은 우리나라의 재생에너지 발전이 전력계통한계가격(System Marginal Price: SMP)에 미치는 영향을 실증적으로 분석한다. 2016~2020년 시간별 데이터를 사용하여 일일 평균, 피크 시간대, 피크 외 시간대의 SMP에 대해 ARX-GARCHX 모형으로 추정한 결과 총부하, 가스도입가격과 태양광 및 풍력 발전이 중요 결정요인으로 작용하고 있고, 특히 태양광 및 풍력 발전이 증가함에 따라 SMP는 하락하는 급전순위효과(merit order effect)가 존재하는 것을 확인하였다. 특히, 태양광 발전량의 1% 증가는 SMP 수준을 0.005%가량 하락시킬 뿐만 아니라 변동성을 증가시키는 것으로 나타났다. 이러한 결과는 재생에너지 발전이 증가함에 따라 SMP가 하락하여 기존 발전소의 수익성 감소와 재생에너지 투자 감소로 이어질 수 있음을 시사하고, 신재생에너지 보조금이 소매 부분 전기요금에 부담을 가중하더라도 계통한계가격이 낮아져서 이를 상쇄할 수 있음을 의미한다.

• 유통과학연구
• /
• 제14권4호
• /
• pp.117-125
• /
• 2016

Purpose - As consumers' needs for purchasing fresh and safe food have been bigger in Korea, their interest in local food is also growing recently. So, the number of local food stores has been increased from 3 in 2012 to 103 in 2015. Local food stores should operate a business responding consumers' needs in order that local food stores are not to be a one-time fad. Therefore, the purpose of this study is to analyze the characteristics of consumers who use a local food store and provide helpful implications to design a strategy for sustainable growth of local food store. Research design, data, and methodology - In this study, Probit model was used for empirical analysis in order to examine the effect of purchase choice attributes of agricultural products, consumer's satisfaction, and their demographic factors upon the intention to use a local food store. After estimating coefficients of the probit model, marginal effects were calculated as a standard normal, and cumulative distribution is differentiated with respect to explanatory variables. To collect the data, questionnaire survey was carried out with the consumers using the local food store (Youngjin Nonghyup near to Jeonju city located in Jeollabuk-do). Result - The data analysis found that the more consumers are satisfied with local food store, the higher intention they have to use the local food store. In addition, it was known that the factors related to quality of agricultural products and shopping convenience among the purchase choice attributes have a considerable impact on the purchase intention of a local food store. In demographic factors, income was turned out to be an important factor affecting purchase intention of local food. Such a result supports the hypothesis that high income consumers are likely to purchase local food, which is based on the inference that consumers who have a high income tend to pursue wellbeing life. Futhermore, information delivery, through a reputable media source among general factors, was known to play an important role in forming an intention to purchase local food. According to the analysis of marginal effects, probability of purchase intention of a local food store is increased by 11.4%, if a monthly average income of a household is above 4.5 million Won(Korean currency). If purchasing satisfaction with local food stores is high, the probability of purchase intention would be increased by 24.1%. Likewise, such a probability goes up by 8.7%, 5.8%, respectively as an increasing one unit of quality of agricultural products and shopping convenience of local food stores, respectively. Conclusion - For attaining sustainable growth in a local food store, it is considered necessarily to establish a proper store operation system to meet consumers' needs, especially for quality and shopping convenience of local food. Moreover, as it was found that appropriate communication through media source has a positive effect on the intention to use local food store, PR activity seems to be necessary to expand the consumers' demands for local foods.

• 한국해안해양공학회:학술대회논문집
• /
• 한국해안해양공학회 1995년도 정기학술강연회 발표논문 초록집
• /
• pp.16-19
• /
• 1995

Effects of the changes in bottom topography and non-linearity of the western boundary current on the separation position of the Tsushima Warm Current(TWC) are investigated using a primitive equation model in a simplified model domain which consists of a deep ocean, a continental shelf and a marginal sea(Fig. 1). (omitted)

• Journal of the Korean Data and Information Science Society
• /
• 제20권5호
• /
• pp.895-903
• /
• 2009

The penalized partial likelihood based on restricted maximum likelihood method has been widely used for the inference of frailty models. However, the standard-error estimate for frailty parameter estimator can be downwardly biased. In this paper we show that such underestimation can be corrected by using hierarchical likelihood. In particular, the hierarchical likelihood gives a statistically efficient procedure for various random-effect models including frailty models. The proposed method is illustrated via a numerical example and simulation study. The simulation results demonstrate that the corrected standard-error estimate largely improves such bias.

• 무역상무연구
• /
• 제57권
• /
• pp.111-135
• /
• 2013

This study analyzes some important spin-off effects of the provision, UCP600 Article 14(6), through the methodology of the expected utility maximization theory based on the state-contingent commodities model. Some technical implications of this study are as follows. First, the risk-averse beneficiary will choose to present his documents more than 5 days before expiry date by paying a higher risk premium (so-called cure period) for full assurance to cure documentary discrepancies, if expressed economically, he pursues loss reducing activities to the point where the expected marginal product of his activities is less than its marginal cost. Secondly, where the effectiveness of securing cure period is uncertain, the risk-averse beneficiary will choose to present documents just on the expiry date without securing any cure period by paying no risk premium. This study finally suggests the safe harbor standard should be optimal solution only if it is supplemented by the hidden reasonableness standard for balancing the conflicts of interest between beneficiaries and banks.