• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.11 no.3
• /
• pp.1115-1125
• /
• 2010

This study is to examine how POS(perceived organization support) would affect trust in her/his superior and organization, organization commitment and customer orientation. The effect of demographic variables differences on independent and dependent variables was examined. Data for this study was collected by questionnaire from a sample of 277 employees of 9 hospitals in Korea. The findings show that POS is significantly influence trust in her/his superior and organization, organization commitment, and customer orientation. Variables like gender and type of employment have a significant difference. However there is no difference in the type of occupation.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2016.05a
• /
• pp.288-288
• /
• 2016

유역단위 오염부하량 산정에는 SWAT, HSPF 등의 물리적 매개변수 기반 분포형 모형이 주로 사용되고 있으나, 공간분포형 입력자료로 인한 많은 매개변수는 모의 과정을 복잡하게 하며, 보정 과정에 있어 많은 시간과 노력을 요구하는 단점이 있다. 이로 인해 실무에서는 원단위법이나 유량-부하량 관계식과 같은 통계적 분석에 의한 회귀식이 주로 사용되고 있다. 그 중 LOADEST는 회귀식 기반 프로그램으로, 다양한 연구자들에 의해 연구되고 있으나, 수질 모형과의 모의능력을 비교하는 연구는 부족하다. 본 연구에서는 청미천 상류유역을 대상으로 유역특성에 따른 LOADEST 기반 회귀식의 매개변수를 추정하여 오염부하량을 모의하고, SWAT 모형에 의한 오염부하량 모의결과와 비교 평가하고자 한다. 모형의 구동 및 회귀식 매개변수 추정에 필요한 입력 자료는 용인시 백암면 일대에서 2013년부터 2015년까지 모니터링한 수질, 유량 및 기상자료와 지형자료 (토지이용도, 토양도, 수치표고자료)를 이용하여 구축하였다. LOADEST 기반 회귀식의 매개 변수 추정은 김계웅 (2015)이 개발한 방법을 사용하였으며, 유역면적, 토지이용비율 등은 지형자료를 이용하여 산정하였다. SWAT 모형의 보정은 2013년부터 2014년까지의 자료를 이용하였으며, 2015년 자료를 이용하여 검정하였다. 본 연구의 결과는 비점오염원 모델에 대한 이해를 넓히고, 오염부하량 모의를 위한 모형 선정에 있어 도움이 될 수 있을 것으로 기대한다.

• The Korean Journal of Applied Statistics
• /
• v.35 no.3
• /
• pp.335-346
• /
• 2022

Profile control chart aims to detect a change in the functional relationship of multivariate characteristics in the statistical process control. In monitoring two variables, a linear profile is of interest composed of the intercept and slope of one variable (response variable) against the other (explanatory variable). The previous studies on monitoring of the linear profile mostly assume that the explanatory variables are the same for all profiles. However, there are also cases where they vary depending on profiles. This paper intends to extend the monitoring method to where explanatory variables are different for each profile. We compare the new method's performance through simulation and apply it to monitoring a network intrusion using NSL-KDD data.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.14 no.6
• /
• pp.1679-1688
• /
• 1990

The method of nonlinear parameter estimation(NPE), which is a statistical and an inverse method, is used to estimate the thermal diffusivity of the porous insulation material. In order to apply the NPE method for measuring the thermal diffusivity, and algorithm for programing suitable to IBM personal computer is established, and is studied the statistical treatment of experimental data and theory of estimation. The experimental data obtained by discrete measurement using a constant heat flux technique are used to find the boundary conditions, initial conditions, and the thermal diffusivity, and then the final values are compared with the values obtained by some different methods. The results are presented as follows:(1) NPE method is used to establish the estimation of the thermal diffusivity and compared results with experimental output shows, that this method can be applicable to define the thermal diffusivity without considering hear flux types. (2) Because of all of the temperatures obtained by the discrete measurement on each steps of time are used to estimate the thermal diffusivity. Although some error in the temperature measurements of temperature are included in estimating process, its influences on the final value are minimzed in NPE method. (3) NPE method can reduce the experimental time including the time of data collecting in a few minutes and can take smaller specimen compared with steady state method. If the tube-type furnace is used, also the adjusting time of surrounding temperature can be reduced.

• Journal of Labour Economics
• /
• v.29 no.2
• /
• pp.67-116
• /
• 2006

This study directly analyzes the wage distributions rather than indirectly looking at a few of their moments. It also investigates wage distributions using various descriptive and semi-parametric methods. The wage distributions of Korean manufacturing industries can in general be represented by three distinct forms, underdeveloped, advanced and the medium of the two. The discrepancies in these distribution forms are explained by differences in the labor-type distributions and their weights in the composition of wage distribution forms, and further clarified through various descriptive statistics based on them. However, the descriptive statistical analysis has a limit in that it shows mixed outcomes of different categoric variables. Then, this problem is resolved by applying a semi-parametric estimation of hazard function and the marginal effect evaluations of variable changes on estimated distributions not on the function. As a result of this marginal analysis, the common features and differences of categoric variables and their intensities of effects on distributions are revealed.

• Korean Business Review
• /
• v.14
• /
• pp.129-148
• /
• 2001

The purpose of this study is to survey the LOV(List Of Values) composed with the consumer's psychographic variables and is to segment consumer's value system through the LOV. Referring to the previous studies such as Rokeach's RVS, Kahke's LOV, Razzaque's Asian values, and author's Korean values, this study conducted researches of consumer's value segmentation. The result of this study finds that the Korean consumer's value system is categorized into (1) value of influence, (2) value of mammonism, (3) value of latest, (4) value of venture, (5) value of salvation, (6) value of sociality, (7) value of curiosity, (8) spirituality, and (9) value of conservativeness in order. This results will help contribute toward segmentation of Korean consumer's value. Another result of this study shows by way of ANOVA(analysis of variance) that the Korean consumer's value dimensions are closely related to demographic variables: sex, marital status, education, occupation, total families income by monthly, family structure, and age.

• Journal of the Korean Geophysical Society
• /
• v.3 no.4
• /
• pp.261-268
• /
• 2000

Minimum magnitudes $(M_c)$ for estimation of seismicity parameters were analyzed for the earthquake catalog of Korea Meteorological Agency (KMA). The temporal variation of earthquake frequency suggests that a proper $M_c$ be 3.0 for the whole southern part of the Korean Peninsula. The b-value with $M_c$ of 3.0 is estimated to be 1.11, which is larger than those of the previous studies. To see the spatial variation of $M_c$, the southern part of the peninsula were divided into grids of $0.1{\times}0.1$ degree. At the greater portion of grid points, the local earthquake catalogs do not satisfy given statistical criteria. The grid points whose local earthquake catalogs meet the criteria mostly distribute in the eastern part. $M_c$ at these points range 2.4 to 3.5 and b values range 0.75 to 1.73 with the average of 1.08 which is comparable to that for the whole southern part of the peninsula.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.20 no.3
• /
• pp.61-68
• /
• 1983

In this paper various aspects of statistical maltiplexing of data signs have been investigated. A queueing model with finite waiting room and batch poisson arrivals is studied assuming that data signals are transmitted at a constant rate. Using traffic intensity and average burst length as parameters, overflow probabilities and expected queueing delay due to buffering are obtained. Also, a real system model of a statistical multiplexer that can be directly used in micro-programmed hardware realization is proposed. To examine the performance of the system, computer simulation has been done at various conditions. The results obtained can be used in designing a buffer efficiently.

• Journal of Korea Water Resources Association
• /
• v.35 no.2
• /
• pp.187-194
• /
• 2002

To improve the accuracy of the areal rainfall estimates over a river basin, the optimal design method of rainfall network was studied using the stochastic characteristics of measured rainfall data. The objective function was constructed with the estimation error of areal rainfall and observation cost of point rainfall and the observation sites with minimum objective function value were selected as the optimal network. As a stochastic variance estimator, kriging model was selected to minimize the error terms. The annual operation cost including the installation cost was considered as the cost terms and an accuracy equivalent parameter was used to combine the error and cost terms. The optimal design method of rainfall network was studied in the Yongdam dam basin whose raingauge numbers need to be enlarged for the optimal rainfall networks of the basin.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2012.05a
• /
• pp.343-343
• /
• 2012

미래 극한사상의 초과확률을 산정하기 위하여 저해상도의 전지구 기후변화 시나리오 자료를 그대로 사용하거나 이를 역학적 또는 통계적 방법으로 상세화한 고해상도 기후변화 시나리오 자료를 활용한다. 통계적 상세화는 전지구 또는 지역기후모델의 현재기후 모의 자료와 관측 자료와의 통계적 관계를 미래 예측자료에 적용하는 방법으로, 현재와 미래 기후의 시공간적 분포가 동일하다는 가정을 포함하고 있다. 반면 역학적 상세화 방법은 기후변화 강제력을 고려하는 지역기후모델을 이용하여 기후시스템의 역학 및 물리과정, 기후시스템간 의 상호작용, 기후변화의 비정상성 등을 고려할 수 있고, 변수간의 시공간적 상관성을 지구시스템의 물리 역학적 과정으로 해석할 수 있다는 장점이 있다. 이에 국립기상연구소에서는 영국 기상청의 통합모델(UM)기반의 지역기후모델(HadGEM3)을 사용하여 50 km 및 12.5 km 격자 단위로 역학적 상세화(dynamic downscaling)를 수행하였다. 본 연구에서는 역학적 상세화로 생산된 HadGEM3-RA 자료를 이용하여 현재기후(1980-2005), 가까운 미래(2020-2049)와 21세기말(2070-2099)의 20년 빈도 강수량을 비교하였다. 연구결과, 남한에 걸쳐 현재기후에 비하여 미래에는 극한강수의 크기와 빈도가 전반적으로 증가하는 경향을 확인할 수 있었다. 20년에 한번씩 발생하였던 일 극한강수는 RCP8.5를 고려한 21세기말에는 약 4년에 한번씩 발생하리라 전망되었다.