• Journal of the Korean Data and Information Science Society
• /
• 제19권2호
• /
• pp.421-429
• /
• 2008

We consider a general semiparametric additive risk model that consists of three components. They are parametric, purely and smoothly nonparametric components. In parametric component, time dependent term is known up to proportional constant. In purely nonparametric component, time dependent term is an unknown function, and time dependent term in smoothly nonparametric component is an unknown but smoothly function. As an estimation method of this model, we use the weighted least square estimation by Huffer and McKeague (1991). We provide an illustrative example as well as a simulation study that compares the performance of our method with the ordinary least square method.

• Communications for Statistical Applications and Methods
• /
• 제29권5호
• /
• pp.561-589
• /
• 2022

When multiple classifications and regression trees are combined, tree-based ensemble models, such as random forest (RF) and Bayesian additive regression trees (BART), are produced. We compare the model structures and performances of various ensemble models for regression settings in this study. RF learns bootstrapped samples and selects a splitting variable from predictors gathered at each node. The BART model is specified as the sum of trees and is calculated using the Bayesian backfitting algorithm. Throughout the extensive simulation studies, the strengths and drawbacks of the two methods in the presence of missing data, high-dimensional data, or highly correlated data are investigated. In the presence of missing data, BART performs well in general, whereas RF provides adequate coverage. The BART outperforms in high dimensional, highly correlated data. However, in all of the scenarios considered, the RF has a shorter computation time. The performance of the two methods is also compared using two real data sets that represent the aforementioned situations, and the same conclusion is reached.

• Asian Pacific Journal of Cancer Prevention
• /
• 제13권4호
• /
• pp.1145-1149
• /
• 2012

Aim: To clarify any association between the hOGG1 Ser326Cys polymorphism and susceptibility to gastric cancer. Methods: A meta-analysis based on 11 eligible case-control studies involving 5,107 subjects was carried out to summarize the data on the association between hOGG1 Ser326Cys polymorphism and gastric cancer risk. Results: No association was found between hOGG1 Ser326Cys polymorphism and gastric cancer risk (dominant model: OR = 0.95, 95% CI: 0.83-1.09, p = 0.486, ph (p values for heterogeneity) = 0.419; additive model: OR = 1.02, 95% CI: 0.81-1.30, p = 0.850, ph = 0.181; recessive model: OR = 1.09, 95% CI: 0.80-1.48, p = 0.586, ph = 0.053). Subgroup analysis based on ethnicity (Asian and Caucasian) and smoking status (ever smoker and never smoker) did did notpresent any significant association. Sensitivity analysis did not perturb the results. Conclusions: This study strongly suggested there might be no association between the hOGG1 Ser326Cys polymorphism and gastric cancer risk. However, larger scale studies are needed for confirmation.

• Journal of Preventive Medicine and Public Health
• /
• 제32권2호
• /
• pp.191-199
• /
• 1999

대기오염이 인체에 미치는 영향을 정량적으로 분석하기 위하여 1991년부터 1995년까지의 서울시 사망자료와 대기오염 자료를 general additive Poisson model 을 사용하여 분석하였다. 우리 나라에서도 외국의 연구 결과와 마찬가지로 대기 오염 물질의 증가가 사망자 수의 증가에 유의하게 영향을 미치는 것으로 평가되었다. 특히 본 연구에서 나타난 특성은 일반적으로 많이 알려진 분진에 의한 영향 보다는 오존에 의한 영향이 두드러진다는 것이다. 오존은 100ppb가 증가할 때 마다 사망자 수를 3-10% 증가시키는 것으로 평가되었다. 즉 오존주의보가 발령될 때마다 서울시에서는 많게는 평소보다 10%정도의 추가적인 사망이 발생하는 것으로 추정된다. 이러한 연구결과는 우리 나라의 대기오염 양상이 분진과 $SO_2$ 중심의 '런던스모그'형에서 오존과 $NO_2$ 등에 의한 광화학적 스모그로 바뀌고 있는 사실을 감안할 때 시사하는 바가 매우 크다고 할 수 있다. 따라서 앞으로는 병원입원율, 또는 응급실 내원환자수에 대한 오존의 영향을 평가할 수 있는 연구들이 이어져야 할 것으로 생각되며, 더불어 오존 오염을 감소시킬 수 있는 구체적인 정책들이 시급히 시행되어야 할 것으로 생각된다.

• 한국원자력학회:학술대회논문집
• /
• 한국원자력학회 1998년도 춘계학술발표회논문집(1)
• /
• pp.483-488
• /
• 1998

A pressure drop model for the PWR grids with and without mixing device is proposed at single phase based on the fluid mechanistic approach. Total pressure loss is expressed in additive way for form and frictional losses. The general friction factor correlations and form drag coefficients available in the open literatures are used to the model. As the results, the model shows better predictions than the existing ones for the non-mixing grids, and reasonable agreements with the available experimental data for mixing grids. Therefore it is concluded that the proposed model for pressure drop can provide sufficiently good approximation for grid optimization and design calculation in advanced grid development.

• Journal of the Korean Data and Information Science Society
• /
• 제27권5호
• /
• pp.1215-1224
• /
• 2016

고해상도 격자 단위 기후정보는 농업, 관광학, 생태학, 질병학 등 다양한 분야의 현상을 설명하는 중요 요인이다. 고해상도 기후정보는 동적 모형과 통계적 모형을 통해 얻을 수 있다. 통계적 모형은 동적 모형에 비해 계산 시간이 저렴하여 시공간 해상도가 높은 기후자료 생성에 주로 이용한다. 본 연구에서는 2003년부터 2012년까지 1월에 관측된 일 평균기온자료를 토대로 통계적 모형의 일 평균 기온을 생성하였다. 통계적 모형으로 선형모형을 기반으로한 일반선형모형, 일반화가법모형, 공간선형모형, 베이지안공간선형모형을 고려하였다. 예측성능평가를 위해 60개소의 지상관측소에서 관측된 일 평균기온을 모형적합 자료로 사용하여 352개소의 자동기상관측의 일 평균기온을 검증하였다. 평균제곱오차와 상관계수를 보면 베이지안공간모형의 예측성능이 다른 모형에 비해 상대적으로 우수하였다. 최종적으로 $1km{\times}1km$ 격자 단위 일 평균기온 지도를 생성하였다.

• 환경영향평가
• /
• 제13권1호
• /
• pp.9-19
• /
• 2004

This study is performed to examine the relationship between air pollution exposure and mortality in Daejeon for the years of 1998 - 2001. Daily counts of death were analyzed by general additive Poisson model, with adjustment for effects of seasonal trend, air temperature, humidity, and day of the week as confounders in a nonparametric approach. Daily death counts were associated with CO(4 day before), $O_3$(current day), $PM_10$(4 day before), $NO_2$(6 day before), $SO_2$(2 day before). Increase of $31.07{\mu}g/m^3$(interquartile range) in $PM_10$ was associated with 2.0 % (95% CI = 0.5 % - 3.5 %)) increase in the daily number of death. This effect was greater in children(less than 15 aged) and elderly(more than 65 aged). We concluded that Daejeon had 2 - 4 % increase in mortality in association with IQR in air pollutants. Daily variations in air pollution within the range currently occurring in Daejeon might have an adverse effect on daily mortality. These findings also support the hypothesis that air pollution at levels below the current ambient air quality standards of Korea except PM10, is harmful to sensitive subjects, such as children or elderly.

• 환경위생공학
• /
• 제18권3호통권49호
• /
• pp.89-99
• /
• 2003

This study is peformed to examine the relationship between air pollution exposure and mortality in Incheon for the years of 1998 - 2001. Daily counts of death were analyzed by general additive Poisson model, with adjustment for effects of seasonal trend, air temperature, humidity, and day of the week as confounders in a nonparametric approach. Daily death counts were associated with CO(1 day before), O$_3$(2 day before), PM$_{10}$(1 day before), NO$_2$(1day before), SO$_2$(1 day before). Increase of 32.21 ${\mu}$g/m$^3$(interquartile range) in PM$_{10}$ was associated with 1.9 % (95% CI = 0.8 % - 2.9 %) increase in the daily number of death. This effect was greater in children(less than 15 aged) and elderly(more than 65 aged). We concluded that Incheon had 2 - 4 % increase in mortality in association with IQR in air pollutants. Daily variations in air pollution within the range currently occurring in Incheon might have an adverse effect on daily mortality. These findings also support the hypothesis that air pollution, at levels below the current ambient air quality standards of Korea, is harmful to sensitive subjects, such as children or elderly.

• Journal of Communications and Networks
• /
• 제14권3호
• /
• pp.225-229
• /
• 2012

We examine spectrum sensing in a situation of uncertain channel model. In particular, we assume that, besides additive noise, the observed signal contains an interference term whose probability distribution is unknown, and only its range and maximum power are known. We discuss the evaluation of the detector performance and its design in this situation. Although this paper specifically deals with the design of spectrum sensors, its scope is wider, as the applicability of its results extends to a general class of problems that may arise in the design of receivers whenever there is uncertainty about how to model the environment in which one is expected to operate. The theory expounded here allows one to determine the performance of a receiver, by combining the available (objective) probabilistic information with (subjective) information describing the designer's attitude.

• Advances in nano research
• /
• 제14권4호
• /
• pp.375-389
• /
• 2023

In this study, free vibration analysis of functionally graded (FG) porous truncated conical shell panels reinforced by graphene platelets (GPLs) has been investigated for the first time. Additionally, the effect of three different types of porosity distribution and five different types of GPLs patterns on dynamic response of the shell are also studied. Halpin-Tsai micromechanical model and Voigt's rule are used to determine Young modulus, shear modulus and Poisson's ratio with mass densities of the shell, respectively. The main novelties of present study are: applying 3D elasticity theory and the finite element method in conjunction with Rayleigh-Ritz method to give more accurate results unlike other simplified shell theories, and also presenting a general 3D solution in cylindrical coordinate system that can be used for analyses of different structures such as circular, annular and annular sector plates, cylindrical shells and panels, and conical shells and panels. A convergence study is performed to justify the correctness of the obtained solution and numerical results. The impact of porosity and GPLs patterns, the volume of voids, the weight fraction of graphene nanofillers, semi vertex and span angles of the cone, and various boundary conditions on natural frequencies of the functionally graded panel have been comprehensively studied and discussed. The results show that the most important parameter on dynamic response of FG porous truncated conical panel is the weight fraction of nanofiller and adding 1% weight fraction of nanofiller could increase 57% approximately the amounts of natural frequencies of the shell. Moreover, the porosity distribution has great effect on the value of natural frequency of structure rather than the porosity coefficient.