• 응용통계연구
• /
• 제37권3호
• /
• pp.283-295
• /
• 2024

무응답을 적절히 처리하기 위한 많은 방법이 연구되었다. 최근 다수의 무응답 대체법이 개발되고 실질적으로 사용되고 있다. 기존에 발표된 다수의 방법은 MCAR (missing completely at random) 또는 MAR (missing at random) 가정을 사용하고 있다. 그러나 관심변수에 영향을 받는 MNAR (missing not at random) 또는 무시할 수 없는 무응답(non-ignorable non-response; NN)은 편향을 발생시켜 대체 결과의 정확성을 크게 떨어뜨리지만 이에 관한 연구는 상대적으로 미미하다. Lee와 Shin (2022)은 등분산 가정하에서 무시할 수 없는 무응답을 적절히 처리할 수 있는 편향수정 무응답 대체법을 제안하였다. 본 연구에서는 Lee와 Shin (2022)이 제안한 방법을 확장한 무응답 대체법으로 초모집단 모형의 오차가 이분산인 경우에서 편향을 제거함으로써 추정의 정확성을 향상하는 방법을 제안하였다. 모의실험을 이용하여 제안된 방법의 타당성을 확인하였다.

• Structural Engineering and Mechanics
• /
• 제18권2호
• /
• pp.137-150
• /
• 2004

The optimization of active bars' placement and feedback gains of closed loop control system for random intelligent truss structures under non-stationary random excitation is presented. Firstly, the optimal mathematical model with the reliability constraints on the mean square value of structural dynamic displacement and stress response are built based on the maximization of dissipation energy due to control action. In which not only the randomness of the physics parameters of structural materials, geometric dimensions and structural damping are considered simultaneously, but also the applied force are considered as non-stationary random excitation. Then, the numerical characteristics of the stationary random responses of random intelligent structure are developed. Finally, the rationality and validity of the presented model are demonstrated by an engineering example and some useful conclusions are obtained.

• 응용통계연구
• /
• 제28권6호
• /
• pp.1113-1120
• /
• 2015

본 논문에서는 사회적으로나 개인적으로 매우 민감한 조사에서 세대별, 연령별 또는 계층별에 따라 조사하고자 하는 모집단이 여러 개의 층으로 구성되어 있고, 각 층이 다지속성으로 되어 있는 경우에, Abul-Ela 등의 다지모형과 Eriksson의 다지무관모형에서 사용한 단순임의추출법 대신에 층화추출법을 적용하여 각 층의 다지속성에 대한 모비율의 추정뿐만 아니라 모집단 전체 모비율에 대한 추정을 할 수 있는 층화 다지 확률화응답모형을 제안하였다. 그리고 층화 다지모형에 있어서 각 층의 표본배분에 대하여 비례배분과 최적배분을 고려하여 다루었다. 또한 층화 다지 확률화응답모형들간의 효율성을 비교해 본 결과 Eriksson의 다지무관모형이 Abul-Ela 등의 다지모형보다 효율적임을 알 수 있었다.

• Steel and Composite Structures
• /
• 제22권5호
• /
• pp.1039-1054
• /
• 2016

This paper presents the dynamic characteristics analysis of the purlin-sheet roofs by the random vibration theories. Results show that the natural vibration frequency of the purlin-sheet roof is low, while the frequencies and mode distributions are very intensive. The random vibration theory should be used for the dynamic characteristics of the roof structures due to complex vibration response. Among the first 20th vibration modes, the first vibration mode is mainly the deformations of purlins, while the rest modes are the overall deformations of the roof. In the following 30th modes, it mainly performs unilateral local deformations of the roof. The frequency distribution of the first 20th modes varies significantly while those of the following 30th modes are relatively sensitive. For different parts, the contributions of vibration modes on the vibration response are different. For the part far from the roof ridge, only considering the first 5th modes can reflect the wind-induced vibration response. For the part near the ridge, at least the first 12 modes should be considered, due to complex vibration response. The wind vibration coefficients of the upwind side are slightly higher than that of the leeward side. Finally, the corresponding wind vibration coefficient for the purlin-sheet roof is proposed.

• Journal of the Korean Data and Information Science Society
• /
• 제17권1호
• /
• pp.187-193
• /
• 2006

This paper shows how to use continuation-ratio logits for the analysis of structured polytomous data. Here, response categories are considered to have a nested binary structure. Thus, conditionally nested binary random variables can be defined in each step. Two types of factors are considered as independent variables affecting response probabilities. For the purpose of analyzing categorical data with binary nested strutures a continuation-ratio mixed model is suggested. Estimation procedure for the unknown parameters in a suggested model is also discussed in detail by an example.

• Journal of the Korean Data and Information Science Society
• /
• 제15권2호
• /
• pp.339-345
• /
• 2004

This paper deals with a mixed logit model for ordered polytomous data. There are two types of factors affecting the response varable in this paper. One is a fixed factor with finite quantitative levels and the other is a random factor coming from an experimental structure such as a randomized complete block design. It is discussed how to set up the model for analyzing ordered polytomous data and illustrated how to estimate the paramers in the given model.

• 대한토목학회논문집
• /
• 제9권4호
• /
• pp.51-61
• /
• 1989

비선형 거동을 일으키는 구조물의 불규칙 진동해석이나 운동방정식이 매개여기와 비제차 여기를 포함하는 능동장비의 동적해석에는 불규칙 진동분야에서 보편적으로 사용되는 스펙트럼해법의 적용이 불가능하다. 만약 불규칙 입력하중을 인공적으로 시뮬레이션 할 수 있다면 이러한 문제들은 운동방정식에 직접적분법을 적용하여 동적응답을 얻을 수 있다. 본 논문에서는 FFT기법을 이용하여 비정상 불규칙 지반가속도를 시뮬레이션하는 효과적이고 실제적인 방법을 제시하며, 이렇게 얻은 인공지진을 응용하는 문제로서 지진하충을 받는 회전축-베어링계의 동적응답을 시뮬레이션하였다. 또한 시뮬레이션기법의 적용성 및 실용성을 검토하기 위한 민감도에 대한 연구도 제시하였다.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 1993년도 가을 학술발표회논문집
• /
• pp.248-255
• /
• 1993

Recently, due to the increasing attention to the structural safety under uncertain environments, many researches on the structural reliability analysis have been peformed. Some useful methods are available to evaluate performance reliability of structures with explicit limit states. However, for large structures, in which structural behaviors can be analyzed with finite element models and the limit states are only expressed implicitly, Monte-Carlo simulation method has been mainly used. However, Monte-Carlo simulation method spends too much computational time on repetitive structural analysis. Many alternative methods are suggested to reduce the computational work required in Monte-Carlo simulation. Response surface method is widely used to improve the efficiency of structural reliability analysis. Response surface method is based on the concept of approximating simple polynomial function of basic random variables for the limit state which is not easily expressed in explicit forms of design random variables. The response surface method has simple algorithm. However, the accuracy of results highly depends on how properly the stochastic characteristics of the original limit state has been represented by approximated function, In this study, an improved response surface method is proposed in which the sampling points for creating response surface are modified to represent the failure surface more adequately and the combined use of a linear response surface function and Rackwitz-Fiessler method has been employed. The method is found to be more effective and efficient than previous response surface methods. In addition more consistent convergence is achieved, Accuracy of the proposed method has been investigated through example.

• Earthquakes and Structures
• /
• 제2권3호
• /
• pp.279-295
• /
• 2011

This paper develops a probabilistic methodology for the seismic reliability analysis of structures with random properties. The earthquake loading is assumed to be described in terms of response spectra. The proposed methodology takes advantage of the response spectra and thus does not require explicit dynamic analysis of the actual structure. Uncertainties in the structural properties (e.g. member cross-sections, modulus of elasticity, member strengths, mass and damping) as well as in the seismic load (due to uncertainty associated with the earthquake load specification) are considered. The structural reliability is estimated by determining the failure probability or the reliability index associated with a performance function that defines safe and unsafe domains. The structural failure is estimated using a performance function that evaluates whether the maximum displacement has been exceeded. Numerical illustrations of reliability analysis of elastic and elastic-plastic single-story frame structures are presented first. The extension of the proposed method to elastic multi-degree-of-freedom uncertain structures is also studied and a solved example is provided.

• 기술혁신학회지
• /
• 제5권2호
• /
• pp.228-244
• /
• 2002

The survey of research and development (R&D) in science and technology (S&T) covers the current status of R&D activities in S&T in Korea, and provides a basis for decision making regarding S&T policy. Continuous improvement of the survey is widely needed to present reliable national basic statistics. Therefore, the purpose of the study is two-fold: to introduce sampling survey method in industrial sector and to make statistical technique to deal with non-response data from industrial sector. To these ends, first, case studies of the United States and Japan are illustrated. A new sampling design for the R&D survey is proposed and implementing stratified random sampling scheme is suggested. Moreover, statistical analysis of the non-response data is dealt with. Based on several screening criteria, we develop a new imputation method suitable for the R&D survey and also provide more detailed implementation plan. Various solutions to a problem arising from non-response item are also presented. Finally, some implications of the results are discussed.