• Communications for Statistical Applications and Methods
• /
• v.9 no.2
• /
• pp.305-313
• /
• 2002

Regarding to inference about a scalar measure of internal scatter of Ρ-variate normal population, this paper considers an interval estimation of the generalized variance, │$\Sigma$│. Due to complicate sampling distribution, fully parametric frequentist approach for the interval estimation is not available and thus Bayesian method is pursued to calculate the highest probability density (HPD) interval for the generalized variance. It is seen that the marginal posterior distribution of the generalized variance is intractable, and hence a weighted Monte Carlo method, a variant of Chen and Shao (1999) method, is developed to calculate the HPD interval of the generalized variance. Necessary theories involved in the method and computation are provided. Finally, a simulation study is given to illustrate and examine the proposed method.

• The Korean Journal of Applied Statistics
• /
• v.25 no.5
• /
• pp.707-718
• /
• 2012

We propose a semi-parametric model to analyze clustered and interval-censored data; in addition, we plugged-in a gamma frailty to the model to measure the association of members within the same cluster. We propose an estimation procedure based on EM algorithm. Simulation results showed that our estimation procedure may result in unbiased estimates. The standard error is smaller than expected and provides conservative results to estimate the coverage rate; however, this trend gradually disappeared as the number of members in the same cluster increased. In addition, our proposed method was illustrated with data taken from diabetic retinopathy studies to evaluate the effectiveness of laser photocoagulation in delaying or preventing the onset of blindness in individuals with diabetic retinopathy.

• Journal of ILASS-Korea
• /
• v.12 no.4
• /
• pp.191-197
• /
• 2007

This study presents in-line digital particle holography and its application to spray droplets to measure the characteristics of spray droplets. Several important parameters at the time of hologram recording such as the object distance and the region of laser beam used were verified. The correlation coefficient method with important parameters such as the reconstruction interval and the correlation interval was used for determination of the focal planes of particles. The optimal values of all these parameters are obtained by either numerical simulation of holograms or experiments. Using these optimal parameters, double pulse digital spray holograms in a short time interval were recorded with the synchronization system for the time control. The spatial positions of droplets that are used for the evaluation of the three dimensional droplet velocities can be easily located, which proves the feasibility of the digital holographic technology for measurements of several important features of spray droplets.

• The Journal of Information Systems
• /
• v.29 no.1
• /
• pp.289-301
• /
• 2020

Purpose This study analyzes whether Internet search of cryptocurrency has a causal relationship to return and volatility of cryptocurrency. Design/methodology/approach Google Trend was used as a measure of the level of Internet search, and the parametric tests of Granger causality in the 1st moment and the 2nd moment were adopted as the analysis method. We used Bitcoin's dollar-based price, which is the No. 1 market value among cryptocurrency. Findings The results showed that the Internet search measured by Google Trends has a causal relationship to cryptocurrency in both average and volatility, while there is a difference in causality and its degree according to the search area and category that Google Trend user should set. Because the Granger causality is based on the improvement of prediction, the analysis results of this study indicate that Internet search can be used as a leading indicator in predicting return and volatility of cryptocurrency.

• Proceedings of the Korea Concrete Institute Conference
• /
• 1997.04a
• /
• pp.422-429
• /
• 1997

Thermal stress induced by hydration heat may produce cracks in mass concrete structure, which can result in structural problems as well as bad appearance. To minimize crack occurrence in massive structural, thus, the study put an emphasis on the determination of optimized lift height and block size. In the parametric study different sizes and lift heights were used to measure the magnitudes of hydration heat and thermal stresses for 3 different types of concrete fabricated with 1 pure cement and 2 blended Portland cements. As a result of analysis. it was found that magnitude of hydration heat and the occurrence of thermal cracks depend on the restriction conditions and material characteristics, especially adiabatic material parameters. It was also found that optimized lift height and block size can be determined from an appropriate combination of the degree of inner and outer structural restrictions.

• Journal of Ocean Engineering and Technology
• /
• v.18 no.1
• /
• pp.22-27
• /
• 2004

An orthotropic steel deck system is widely adapted form for a long-span bridge. It has many advantages, such as the big reduction of dead weight, the simplicity for erection, and the reduction of the construction period. However, an orthotropic steel deck system requires a lot of welding work, which may result in defects and deformation of connection. Therefore, the research for the general behavior and fatigue strength of the several details in orthotropic steel deck bridge is necessary. The fatigue failure with distortion results from secondary stress by out-of-plane deformation; these kinds of cracks are very difficult to measure, and can not be precisely calculated through finite element analysis. This stress concentration phenomenon generates the fatigue failure around the lower scallop of the transverse rib. This paper presents improved details of the intersection between the longitudinal rib and the transverse rib of an orthotropic steel deck bridge by the third dimensional hit size test, and the finite element method, which can minimize local stress through parametric study.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.53 no.5
• /
• pp.249-256
• /
• 2004

This paper suggests a novel evaluation scheme of protective effectiveness in distribution systems. The adequacy of every parameter in a protective device is evaluated for the setting or correction rules. Then, the protective effectiveness of a device, device-wise effectiveness, is obtained by the combination of the parametric evaluation results. The coordination-wise effectiveness between devices can be calculated by evaluating the parameters which contribute the performance of coordination. The protective effectiveness of the whole system can be obtained by combining the device-wise and coordination-wise effectiveness values. The rules, in this paper, are categorized into three groups; rules for single parameter, rules for coordination between parameters, and rules for coordination between protective devices to form a hierarchical calculation model. The proposed method is applied to a typical distribution network to show its effectiveness.

• Earthquakes and Structures
• /
• v.2 no.1
• /
• pp.1-24
• /
• 2011

During very strong earthquakes, seismically isolated buildings may experience large horizontal relative displacements, which may lead to poundings if an insufficiently wide clearance is provided around the building. This paper investigates, through numerical simulations, the effectiveness of using rubber bumpers, which could be attached at locations where it is likely to have impacts, in order to act as shock-absorbers. For the simulation of the dynamic behavior of such rubber bumpers during impacts, a nonlinear force-based impact model, which takes into account the finite thickness of the rubber bumpers, has been developed. Subsequently, a series of parametric analyses are performed to assess the effect of the gap size, the earthquake characteristics and the thickness, compressive capacity and damping of the bumpers. The stiffness of the moat wall is also parametrically considered during poundings of a seismically isolated building, as another potential mitigation measure for poundings of seismically isolated buildings.

• Journal of the Korean Society of Safety
• /
• v.28 no.1
• /
• pp.74-80
• /
• 2013

This study describes structural reliability analysis of actively-controlled structure for which random vibration analysis is incorporated into the first-order reliability method (FORM) framework. The existing approaches perform the reliability analysis based on the RMS response, whereas the proposed study uses the peak response for the reliability analysis. Therefore, the proposed approach provides us a meaningful performance measure of the active control system, i.e., realistic failure probability. In addition, it can deal with the uncertainties in the system parameters as well as the excitations in single-loop reliability analysis, whereas the conventional random vibration analysis requires double-loop reliability analysis; one is for the system parameters and the other is for stochastic excitations. The effectiveness of the proposed approach is demonstrated through a numerical example where the proposed approach shows fast and accurate reliability (or inversely failure probability) assessment results of the dynamical active control system against random seismic excitations in the presence of parametric uncertainties of the dynamical structural system.

• Communications for Statistical Applications and Methods
• /
• v.6 no.3
• /
• pp.849-860
• /
• 1999

This paper addresses the issue of estimating regression function with errors in variables using wavelets. We adopt a nonparametric approach in assuming that the regression function has no specific parametric form, To account for errors in covariates deconvolution is involved in the construction of a new class of linear wavelet estimators. using the wavelet characterization of Besov spaces the question of regression estimation with Besov constraint can be reduced to a problem in a space of sequences. Rates of convergence are studied over Besov function classes $B_{spq}$ using $L_2$ error measure. It is shown that the rates of convergence depend on the smoothness s of the regression function and the decay rate of characteristic function of the contaminating error.