• The Korean Journal of Applied Statistics
• /
• v.28 no.1
• /
• pp.61-74
• /
• 2015

Structural equation modeling (SEM) is a widely used in social sciences such as education, business administration, and psychology. In SEM, the latent variable score is the estimate of the latent variable which cannot be observed directly. This study uses stagewise structural equation modeling(stagewise SEM; SSEM) by partitioning the whole model into several stages. The traditional estimation method minimizes the discrepancy function using the variance-covariance of all observed variables. This method can lead to inappropriate situations where exogenous latent variables may be affected by endogenous latent variables. The SSEM approach can avoid such situations and reduce the complexity of the whole SEM in estimating parameters.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.34 no.3
• /
• pp.939-945
• /
• 2014

In most of mobility enhancing plans for the transportation disabled, decisions for the investment priorities are firstly made by the facilities that have lower installation rate or lower satisfaction rate; the decisions are made without analyzing the importance factor (path loading factor) between the facility installation rate and the satisfaction rate together. In this study, a novel method of finding causality between the exogenous latent variables and the endogenous latent variables is provided by using the Structural Equation Model (SEM). The results show that the most influential facilities for the transportation disabled are bus stops, crosswalks and sidewalks in order. Also, a curb height around bus stops, a smoothness of sidewalks and installation of crosswalks traffic light are identified as an important facilities for the Transportation disabled.

• Proceedings of The Korean Society of Health Promotion Conference
• /
• 2010.05a
• /
• pp.148-148
• /
• 2010

본 연구는 서울시 지역주민을 대상으로 스트레스, 우울, 문제음주 간의 인과관계를 검증하고자 시도되었다. 본 연구는 서울시 S구이며, 전체 17개동에서 만 19세 이상인 남녀 주민을 층화집락추출법을 통해 표집된 1234명을 대상으로 하는 면접식 단면조사연구를 수행하였다. 구조방정식 모형을 사용하였으며, 측정모형에서 우울, 알코올장애의 내생잠재변수와 외생잠재변수인 스트레스 및 인구학적 변수(성, 연령, 결혼상태)를 이론변수로 설정하여 이들 간의 상관관계를 추정하였다. 또한 문제음주를 최종내생변인으로 우울을 매개변인으로 설정한 상태에서 경로분석을 실시하여 각 변인들의 상대적 중요성에 대한 평가를 시도하였다. 문제음주에 영향을 미치는 변수는 여성보다는 남성이($\beta=-6.64$, p < .001), 연령이 낮을수록($\beta=-.08$, p < .001), 스트레스가 높을수록($\beta=-.08$, p < .001), 미혼이 기혼($\beta=-1.15$, p < .05)의 순으로 경로계수가 유의하였다. 우울에 미치는 영향력은 스트레스($\beta=-.56$, p < .001)와 문제음주정도($\beta=-.15$, p < .001)의 순으로 경로계수가 유의하였다. 이상과 같이 우울이 직접적으로 문제음주에 영향을 주기보다는 스트레스로 인하여 문제음주가 유발될 수 있고 문제음주를 경유하여 우울에 영향을 미치는 것으로 드러났다. 이상의 연구결과를 종합해 볼 때 문제음주는 스트레스와 우울과 관련성이 크므로 우선적으로 스트레스를 적극적으로 관리하는 것이 중요하다. 또한 알코올과 관련한 통제정책으로 절대적인 음주량을 감소시키는 것이 필요하며, 문제음주자를 조기 발견하여 정신건강에 문제로 발전하지 않도록 하는 예방프로그램의 개발이 시급하다 하겠다.

• Management & Information Systems Review
• /
• v.31 no.4
• /
• pp.469-490
• /
• 2012

This study reviews the issue of endogeneity which has gained in importance in the empirical accounting researches in recent years. In so doing, I aim to call attention to the extent to which the presence of endogeneity limits the validity of empirical testing of models. More specifically, this paper explains what endogeneity is, its causes and consequences, and potential ways of managing the problem. The issues that concern endogeneity include, in the main, omitted variables, simultaneity, equilibrium conditions, choice variables. In this paper, I also discuss the implications derived from several selected empirical accounting research topics, that explicitly dealt with the problem of endogeneity. These include, among others, Oswald and Zarowin's (2007) study on the informativeness of capitalization of R&D, Hazarika et al. (2012) focusing on the relationship among internal corporate governance, CEO turnover, and earnings management, and a series of literature devoted to the associations between corporate governance structure and firm values. Finally, possible approaches for dealing with endogeneity are discussed.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.34 no.1
• /
• pp.223-233
• /
• 2014

This study empirically analyzes the relationships among the road and traffic experts' personal characteristics, the driving behavior and factors being expected to have an effect on the satisfaction about roundabout operation. The factors are drawn and grouped through the principle component analysis to clarify driving environment satisfaction on roundabout operation. Each group is named as personal attribute, driving behavior attribute, and satisfaction. After the variables are refined by confirmatory factor analysis, satisfaction model is developed with personal attribute and driving behavior attributes as exogenous variables and roundabout driving awareness and emotion attributes as endogenous variables. As a result, driving satisfaction of roundabout operation is directly influenced by delay reduction, safety improvement, capacity increase, sight improvement, severity accident reduction, and bicycle convenience and indirectly gender, age, driving time, and driving experience. Law obeyance, driving concession, traffic sign obeyance, and interposition do not statistically shows significant on satisfaction. As a result of Analytical Hierarchy Process (AHP), the turning radius of geometry and the driving behavior are important elements for roundabout safety.

• KDI Journal of Economic Policy
• /
• v.14 no.4
• /
• pp.3-26
• /
• 1992

The pupose of this paper is to construct a price determination model of the Korean economy and to find out the propogation mechanism of monetary and fiscal policies. The model is a small-size macroeconometric model consisted of ten core equations : consumption, investment, exports, imports, consumer price index, wage rate, corporate bond rate, potential GNP, capital stock, and GNP identity. The model is a Keynesian model : consumer price index is determined by markup over costs, and wage rate is expressed by Phillipse curve ralation. Two features of the model, however, distinguish this model from other macroeconometric models of the Korean economy. First of all, the estimation of potential GNP and the capital stock is endogenized as suggested by Haque, Lahiri, and Montiel (1990). This allows us to calculate the level of excess demand, which is defined as the difference between the actual GNP and the potential GNP. Second, interest rate, inflation and wages are all estimated as endogenous variables. Moreover, all quantity variables include price variables as important determinants. For instance, interest rate is an important determinant of consumption and investment. Exports and imports are determined by the real effective exchange rate. These two features make the interactions between excess demand and prices the driving forces of this model. In the model, any shock which affects quantity variable(s) affects excess demand, which in turn affects prices. This strong interaction between prices and quantities makes the model look like a classical model over the long run. That is, increases in money supply, government expenditures, and exchange rate (the price of the U.S. dollar in terms of Korean won) all have expansionery effects on the real GNP in the short run, but prices, wage, and interest rate all increase as a result. Over the long run, higher prices have dampenning effects on output. Therefore the level of real GNP turns out to be not much different from the baseline level ; on the other hand, the rates of inflation, wage and interest rate remain at higher levels.

• Environmental and Resource Economics Review
• /
• v.19 no.3
• /
• pp.597-632
• /
• 2010

There is general agreement that global warming is occurring and that the main contributor to this probably is the buildup of green house gasses, GHG, in the atmosphere. Two main contributors are the utilization of fossil fuels and the deforestation of many regions of the world. The burning of fossil fuels increases atmospheric carbon while the burning of fuel-wood reducing fossil fuel consumption along with its forest source maintain an atmospheric carbon level. The standing timber in the forests is a carbon sink, as are wood buildings and structures, and fossil fuel in the ground. This paper is designed to examine a number of current issues related to mitigating the global warming problem through forestry. For this purpose, we develop a modeling approach by integrating timber market, fossil fuel market and carbon cycling model. We use discrete time optimal control theory to identify optimal time paths, the laws of motion, and stationary stats solutions of endogenous variables in the model. On the basis of these results, we identify the optimal amounts of subsidies to be provided or taxes to be imposed by the regulatory agency to mitigate atmospheric carbon accumulation. We also present a numerical example to help illustrate the characteristics of variables in the model when the social cost for atmospheric carbon incrementally shifts upward. A surprising result is that the social cost function for atmospheric carbon has a very smaller impact on the optimal rotation period than previous literature suggested.