• 응용통계연구
• /
• 제28권1호
• /
• pp.61-74
• /
• 2015

최근 교육학, 경영학, 심리학 등 사회과학 뿐만 아니라 공정관리, 생물정보학 등 자연과학에서도 널리 사용되고 있는 구조식 모형(structural equation modeling)에서 잠재변수점수(latent variable score)는 직접 측정이 불가능한 잠재변수를 수량화한 추정치이다. 이 연구에서는 구조식 모형을 단계(stage)별로 분할하여 분석하는 단계별 구조식 모형(stagewise SEM; SSEM)을 제안하였다. 기존 방법은 모든 관측변수의 분산-공분산을 한꺼번에 고려하므로 독립변수인 외생잠재변수(exogenous latent variable)가 종속변수인 내생잠재변수(endogenous latent variable)에 의해 영향을 받는, 논리적으로 타당하지 않은 경우가 있다. 단계별 구조식 모형은 이런 문제점을 해결할 뿐만 아니라 모형의 복잡성을 낮추어 쉽게 해를 찾을 수 있으며, 분석과정에서 생성되는 잠재변수점수로 추가 분석도 용이하다.

• The Journal of Asian Finance, Economics and Business
• /
• 제9권2호
• /
• pp.121-131
• /
• 2022

The goal of the study was to see how quality management (QMA), strategy (STR), and innovative capability (INC) influence the competitive advantage of a Thai information communication technology (ICT) firm (COA). The researchers collected 431 surveys from Thailand's owners and managers employed in ICT enterprises from the beginning of June 2021 to the end of September 2021using diverse sample strategies. A questionnaire with an index of item-objective congruence (IOC) value of 0.60-1.00 and a reliability value of 0.92-0.96 was used as the research tool. Participants in the survey were requested to fill out a seven-level opinion survey posted on Google Forms. A latent variable structural equation model (SEM) path analysis using LISREL 9.1 was used for the four latent variables, 31 manifest variables, and the five hypotheses testing. The analysis showed that all three causal variables positively affected COA, which had a total effect (TE) R² value = 80% when combined with the other latent variables. Moreover, the values for the latent variables when ranked by total effect (TE) were STR, QMA, and INC with TE values of 0.95, 0.89, and 0.25, respectively. Finally, there were very strong influences from COA to STR (0.95), INC to QMA (0.86), and STR to QMA (0.71).

• 한국데이터정보과학회:학술대회논문집
• /
• 한국데이터정보과학회 2005년도 추계학술대회
• /
• pp.177-188
• /
• 2005

Association rule mining searches for interesting relationships among items in a given large data set. Association rules are frequently used by retail stores to assist in marketing, advertising, floor placement, and inventory control. There are three primary threshold measures in association rule; support and confidence and lift. In the case of appling real world to association rules, we have some difficulties in data interpretation because we obtain many rules. In this paper, we develop the model of association rules using latent variables for environmental survey data.

• Journal of the Korean Data and Information Science Society
• /
• 제17권1호
• /
• pp.149-160
• /
• 2006

Association rule mining searches for interesting relationships among items in a given large data set. Association rules are frequently used by retail stores to assist in marketing, advertising, floor placement, and inventory control. There are three primary threshold measures in association rule; support and confidence and lift. In the case of appling real world to association rules, we have some difficulties in data interpretation because we obtain many rules. In this paper, we develop the model of association rules using latent variables for environmental survey data.

• Nuclear Engineering and Technology
• /
• 제55권10호
• /
• pp.3628-3638
• /
• 2023

This paper uses 135 Licensed Operator Event Reports (LOER) from Chinese nuclear plants to analyze how safety culture affects unsafe behaviors in nuclear power plants. On the basis of a modified human factors analysis and classification system (HFACS) framework, structural equation model (SEM) is used to explore the relationship between latent variables at various levels. Correlation tests such as chi-square test are used to analyze the path from safety culture to unsafe behaviors. The role of latent error is clarified. The results show that the ratio of latent errors to active errors is 3.4:1. The key path linking safety culture weaknesses to unsafe behaviors is Organizational Processes → Inadequate Supervision → Physical/Technical Environment → Skill-based Errors. The most influential factors on the latent variables at each level in the HFACS framework are Organizational Processes, Inadequate Supervision, Physical Environment, and Skill-based Errors.

• Journal of the Korean Statistical Society
• /
• 제23권2호
• /
• pp.463-482
• /
• 1994

In Bayesian analysis of randomized response models, the likelihood function does not combine tractably with typical priors for the parameters of interest, causing computational difficulties in posterior analysis of the parameters of interest. In this article, the difficulties are solved by introducing appropriate latent variables to the model and using the Gibbs sampling algorithm.

• International Journal of Quality Innovation
• /
• 제6권2호
• /
• pp.147-166
• /
• 2005

The purpose of this study is to analyze the causal relationship, in the perspective of Total Quality Management, among the education quality factors, which were suggested in the previous researches. Lee et al. [16] had tried to analyze the relationship among education factors, but they did not estimate the education factor using latent variable concept, which is very reasonable and efficient to represent the constructed concepts. So this study attempts to analyze the causal relationship among education quality factors, represented as latent variables used in structural equation modeling (SEM), and compared with each other. In this study, education quality factors were measured by several measures, constructed as several latent variables, and then processed with AMOS, the most efficient statistical package in the SEM area. In order to analyze the causal relationship among the education quality factors constructed as latent variables, this study designed the structural equation model with suggested factors and established several research hypotheses. This study discovered a prominent causality among the education quality factors, such as education leadership, student scholastic performance and satisfaction of education quality, which is similar to that of previous research. This outcome is really a unique Korean syndrome manifest within our educational career orientation.

• 자원ㆍ환경경제연구
• /
• 제16권3호
• /
• pp.701-721
• /
• 2007

진술선호법(stated preference methods)을 이용하는 비시장가치 연구의 핵심과제는 응답자 이질성(heterogeneity)에 대한 대응이라 해도 과언이 아니다. 기존 연구에서 이를 위해 사용되는 두 가지 접근법은 (1) 사회경제변수와 같은 응답자 특성변수를 간접효용함수에 포함시키는 것과 (2) 변수 추정치의 분산을 허용하는 mixed legit이나 probit과 같은 일반화된 모델을 적용하는 것이다. 본 연구에서는 심리태도변수(attitudinal variables)를 특성변수로서 사용하는 세 번째 접근법을 제시하고 검증하였다. 심리태도 변수는 초이스 모델링 연구에서 설명변수와 분할 지표변수(segmentation criteria)로 적용되었다. 연구결과에 의하면 심리태도변수는 모델의 신뢰성과 속성 변수 추정치뿐만이 아니라 지불의사액에도 영향을 주는 것으로 나타났다.

• 산업공학
• /
• 제12권4호
• /
• pp.506-513
• /
• 1999

We use a path analysis to identify influential latent variables on employees' anticipated commitment to a merged bank. Survey samples are taken from Hanvit bank which is a merged form of Hanil and Sangup. Latent variables used in the path analysis are perceptions of organizational support, contact conditions, organizational unity, employee threat and organizational commitment. We find an interesting pre-merger group membership effect as well as merger pattern perceived by employees on the path for the anticipated commitment to the merged bank.

• Communications for Statistical Applications and Methods
• /
• 제8권3호
• /
• pp.777-786
• /
• 2001

The autoregressive models have been used to describe a wade variety of time series. Then the problem of determining the order in the times series model is very important in data analysis. We consider the Bayesian approach for finding the order of autoregressive(AR) error models using the latent variable which is motivated by Tanner and Wong(1987). The latent variables are combined with the coefficient parameters and the sequential steps are proposed to set up the prior of the latent variables. Markov chain Monte Carlo method(Gibbs sampler and Metropolis-Hasting algorithm) is used in order to overcome the difficulties of Bayesian computations. Three examples including AR(3) error model are presented to illustrate our proposed methodology.