• The Korean Journal of Applied Statistics
• /
• v.18 no.2
• /
• pp.435-443
• /
• 2005

Coefficients of determination in logistic regression analysis are defined as various statistics, and their values are relatively smaller than those for linear regression model. These coefficients of determination are not generally used to evaluate and diagnose logistic regression model. Liao and McGee (2003) proposed two adjusted coefficients of determination which are robust at the addition of inappropriate predictors and the variation of sample size. In this work, these adjusted coefficients of determination are applied to variable selection method for logistic regression model and compared with results of other methods such as the forward selection, backward elimination, stepwise selection, and AIC statistic.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.15 no.1
• /
• pp.39-48
• /
• 2011

In this present study, a response modification factor for a lightweight steel panel-modular system which is not clarified in a current building code was proposed. As a component of the response modification factor, an over-strength factor and a ductility factor were drawn from the nonlinear static analysis curves of the systems modeled on the basis of the performance tests. The final response modification factor was then computed by modifying the previous response modification factor with a MDOF (Multi-Degree-Of-Freedom) base shear modification factor considering the MDOF dynamic behaviors. As a result of computation for the structures designed as a dual frame system, ranging from 2-story to 5-story, the value of 4 was estimated as a final response modification factor for a seismic design, considering the value of 5 as an upper limit of the number of stories.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.8 no.1
• /
• pp.29-37
• /
• 2004

The overstrength factor and the ductility factor are the two important factors that determine response modification factors used in current seismic codes, In this paper the overstrength and ductility factors of special concentric braced frames are determined by performing pushover analysis of model structures with various stories and span lengths. and by using those factors the response modification factors are obtained. According to the analysis results. the overstrength and the ductility factors are larger than the values proposed by the codes in low-rise structures. and the opposite is true in medium to high-rise structures, It is also found that the factors increase as the height of structures decreases and the span length increases.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2012.05a
• /
• pp.227-227
• /
• 2012

본 연구는 수자원계획 문제에서 다기준 의사결정기법을 적용할 때 발생할 수 있는 두 가지 문제에 대해 분석하였다. 첫 번째는 다기준 의사결정기법 선택의 차이가 결과에 어느 정도 영향을 미칠 수 있는지를 제시하였고 두 번째는 평가기준에 대한 가중치와 대안들의 평가치에 대한 불확실성을 최소화하기 위해 민감도 분석을 수행하는 절차를 제시하였다. 첫 번째 문제를 위해 가중합계법, Compromise Programming, 계층화분석과정, 수정된 계층화 분석과정, 가중곱방법, TOPSIS, ELECTRE-2, Regime 방법을 사용하였다. 또한 최근 사용빈도가 높은 삼각형 Fuzzy 숫자와 다기준 의사결정기법을 결합한 기법에 대해서도 분석하였는데 Fuzzy WSM, Fuzzy 계층화분석과정, Fuzzy 수정 계층화분석과정, Fuzzy TOPSIS, Fuzzy Compromise Programming을 검토하였다. 분석결과 평가기준에 대한 가중치 조건과 표준화 방법이 동일한 상황에도 불구하고 조금씩 다른 순위를 제시하는 것으로 나타났다. 또한 다양한 MCDM 기법들을 적용해도 동일한 순위로 나타나는 대안들이 있었다. 따라서 다기준 의사결정기법을 사용한 수자원 관리계획을 수립할 때에는 다양한 분석기법을 활용해서 기법의 선택으로 인한 불확실성을 최소화해야 한다. 두 번째 문제는 평가기준에 대한 가중치와 대안의 효과 정량화 자료의 불확실성을 극복하기 위해 각각에 대한 민감도 분석을 수행하였다. 본 연구는 유량확보와 수질개선을 위한 수자원 계획 수립을 위해 가중합계법을 이용한 문제에 두 경우의 민감도 분석을 모두 수행하였다. 이 과정에서 결정계수와 민감도 계수를 산정하여 이용하였다. 본 연구는 향후 수자원 관리 및 계획 분야에서 다기준 의사결정기법을 적용할 때 사용될 수 있는 기초 가이드라인이 될 것이다.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.26 no.1
• /
• pp.49-57
• /
• 2013

Structural elements of typical bridges are superstructure, connections, substuctures and foundations and earthquake resistance is decided with the failure mechanism formed by substuctures and connections. Therefore earthquake resistant design should be carried out in the basic design step where design strengths, e.g. design sections for structural elements are determined. The Earthquake Resistant Design Part of Korean Roadway Bridge Design Code provides two basic design procedures. The first conventional procedure applies the Code-provided response modification factors. The second new procedure is the ductility-based earthquake resistant design, where designer can determine the response modification factors. In this study, basic designs including the two design processes are carried out for a typical bridge and supplements are identified in view of providing earthquake resistance.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.25 no.2
• /
• pp.149-154
• /
• 2012

While the Earthquake Resistant Design Part of Korean Roadway Bridge Design Code provides design procedures for the No Collapse Requirement, requirements for the Serviceability Limit State are not clearly provided. The basic design method to meet the No Collapse Requirement is the spectrum analysis method using response modification factors and the Serviceability Limit State is determined by both the importance factor and the response modification factor applied in the design procedure. The importance factor can be simply applied according to the bridge importance category, however, in moderate/low seismic regions the application of the response modification factor may bring different result according to design conditions. In this study, for a typical bridge in the moderate/low seismic regions, determination procedures for the Serviceability Limit State are reviewed by carrying out earthquake resistant design and supplementary provisions for the Earthquake Resistant Design Part are identified based on the study results.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2011.05a
• /
• pp.403-407
• /
• 2011

본 연구에서는 임의시간 환산계수(Conversion Factor, CF)와 관련하여 거의 유일한 이론적 연구인 Weiss 모형을 자세히 검토하고, 그 문제점을 보완한 수정 Weiss 모형을 제안하였다. 특히, 임의시간 환산계수를 이론적인 방법을 통해 산정하는 과정에서 강우 지속기간과 강우 시간분포의 특성을 반영하지 못하는 한계를 극복하는 시도의 하나로서 강우의 시간분포 특성을 고려하여 CF를 결정하였다. 이를 실제 관측 자료에 근거한 국내 외 CF값들과 비교해본 결과, 수정 Weiss 모형을 근거로 산정한 CF값은 강우 시간분포의 특성에 영향을 받으며 그 형태에 따라 다양한 값이 나타남을 확인할 수 있었다.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.19 no.2
• /
• pp.463-474
• /
• 1995

Modified Vainshtok weight function method is developed. The thermal shock stress intensity factors for elliptical surface cracks existed in the thin and thick walled cylinders are determined. The present results are compared with previous solutions and shown to be good agreement with them.

• Journal of the Institute of Convergence Signal Processing
• /
• v.4 no.4
• /
• pp.52-57
• /
• 2003

A PID parameter tuning method was presented by the fuzzy singleton inference, based on step response-shaping of plant and experience knowledge of expert. The parameter-tuning has tow levels. The higher level determines modified coefficients for the controller based on operator's tuning know-how for characteristics of plant which can not be modeled. The lower level determines specified coefficients based on characteristics of response by Ziegler-Nickel's bounded sensitivity method. The last level parameters tuning of a PID controller is adjusted which the modified and specified coefficients makes adjustment rule, and is adjusted the proper value to each parameters by fuzzy singleton inference. Moreover, proposed the tuning method can reflex exporter knowledge and operator's tuning know-how and fuzzy singleton inference is rapidly operated.

• The Korean Journal of Applied Statistics
• /
• v.18 no.3
• /
• pp.701-712
• /
• 2005

The suppression for logistic regression models has been debated no longer than that for linear regression models since, among many other reasons, sum of squares for regression (SSR) or coefficient of determination ($R^2$) could be defined into various ways. Based on four kinds of $R^2$'s: two kinds are most preferred, and the other two are proposed by Liao & McGee (2003), four kinds of SSR's are derived so that the suppression for logistic models is explained. Many data fitted to logistic models are generated by Monte Carlo method. We explore when suppression happens, and compare with that for linear regression models.