• Genomics & Informatics
• /
• v.5 no.3
• /
• pp.95-101
• /
• 2007

In this paper, we consider the variable selection methods in the Cox model when a large number of gene expression levels are involved with survival time. Deciding which genes are associated with survival time has been a challenging problem because of the large number of genes and relatively small sample size (n<

• The Journal of Asian Finance, Economics and Business
• /
• v.8 no.3
• /
• pp.527-538
• /
• 2021

The purpose of the study is to determine the efficiency, position, and partner selection of securities companies via the negative super-SBM model used in data envelopment analysis (DEA). This model utilizes a variety of inputs, including current assets, non-current assets, fixed assets, liabilities, owner's equity and charter capital, and outputs including net revenue, gross profit, operating profit, and net profit after tax collected from the financial reports (Vietstock, 2020) of 32 securities companies, operating during the period from 2016 to 2019, negative data are collected as well. Empirical results determined both efficient and inefficient terms, and then further determined the position of each securities firm under consideration of every term. The overall score arrived at discovered a large performance change realizing a maximum score able to reach 20.791. In the next stage, alliancing inefficient companies was carried out based on the 2019 scores to seek out optimal partners for the inefficient companies. The tested result indicated that AAS was the best partner selection when its partners received a good result after alliancing, as with FTS (11.04469). The partner selection is deemed as a solution helpful to inefficient securities companies in order to improve their future efficiency scores.

• Communications for Statistical Applications and Methods
• /
• v.31 no.4
• /
• pp.441-457
• /
• 2024

Semicontinuous data are characterized by a mixture of a point probability mass at zero and a continuous distribution of positive values. This type of data is often modeled using a two-part model where the first part models the probability of dichotomous outcomes -zero or positive- and the second part models the distribution of positive values. Despite the two-part model's popularity, variable selection in this model has not been fully addressed, especially, in high dimensional data. The objective of this study is to investigate variable selection and prediction performance of penalized regression methods in two-part models. The performance of the selected techniques in the two-part model is evaluated via simulation studies. Our findings show that LASSO and ENET tend to select more predictors in the model than SCAD and MCP. Consequently, MCP and SCAD outperform LASSO and ENET for β-specificity, and LASSO and ENET perform better than MCP and SCAD with respect to the mean squared error. We find similar results when applying the penalized regression methods to the prediction of crime incidents using community-based data.

• The Korean Journal of Applied Statistics
• /
• v.34 no.3
• /
• pp.411-425
• /
• 2021

Accelerated failure time (AFT) model represents a linear relationship between the log-survival time and covariates. We are interested in the inference of covariate's effect affecting the variation of survival times in the AFT model. Thus, we need to model the variance as well as the mean of survival times. We call the resulting model mean and variance AFT (MV-AFT) model. In this paper, we propose a variable selection procedure of regression parameters of mean and variance in MV-AFT model using penalized likelihood function. For the variable selection, we study four penalty functions, i.e. least absolute shrinkage and selection operator (LASSO), adaptive lasso (ALASSO), smoothly clipped absolute deviation (SCAD) and hierarchical likelihood (HL). With this procedure we can select important covariates and estimate the regression parameters at the same time. The performance of the proposed method is evaluated using simulation studies. The proposed method is illustrated with a clinical example dataset.

• Journal of Korean Society for Quality Management
• /
• v.32 no.2
• /
• pp.77-92
• /
• 2004

It is difficult but very necessary to measure the productivity of container terminals as logistics service provider. It is meaningful to find the appropriate inputs and outputs of the logistics service delivery systems and to measure the relationship between these inputs and outputs. This study proposes a model of evaluating the efficiency of container terminals. The evaluation consists of three phases. First, DEA(Data Envelopment Analysis) phase, determines the efficiency score and weights of DMUs(Decision Making Unit). This phase performs through four steps : selection of DMU, selection of DEA model, determination of input and output factors, calculation of efficiency score and weights for each DMU. Secondly, CEM (Cross Evaluation Model) phase, is to calculate the cross-efficiency scores of DMUs. This phase performs through three steps: selection of CEM, determination of cross-efficiency score for each DMU and development of cross-efficiency matrix. Finally, average cross-efficiency analysis phase is to compute the average cross-efficiency score. The proposed model discriminates among DMUs and ranks DMUs, whether they are efficient or inefficient.

• The Korean Journal of Applied Statistics
• /
• v.22 no.1
• /
• pp.197-204
• /
• 2009

In the model selection problem, the main objective is to choose the true model from a manageable set of candidate models. An information criterion gauges the validity of a statistical model and judges the balance between goodness-of-fit and parsimony; "how well observed values ran approximate to the true values" and "how much information can be explained by the lower dimensional model" In this study, we introduce some information criteria modified from the Akaike Information Criterion (AIC) and the Bayesian Information Criterion(BIC). The information criteria considered in this study are compared via simulation studies and real application.

• Journal of Korea Port Economic Association
• /
• v.33 no.4
• /
• pp.37-58
• /
• 2017

The implicit assumption of port selection studies based on survey and respondents' perceptions is that the preference of the port selection attributes is proportional to the selection behavior. Further, the straight lines of the port selection attributes could also have non-linear properties. This study confirms nonlinear characteristics of selection attributes by using Kano model. The findings of this study showed that several properties of carriers were evaluated as nonlinear characteristics, such as the intermodal links and network accessibility, and size of port and terminal. Hence, port service providers such as port authorities and terminal operating companiesl, should construct a port operation strategy that reflects the non-linear port selection characteristics of shipping companies. Since this study aimed at exploring the forms of port selection characteristics, long-term additional verification studies on ports and stakeholders at domestics and abroad were needed. The Kano model and importance-selection analysis method used for analysis and strategy establishment also need to be improved to capture evident characteristics and to present strategic guidelines.

• Journal of Korean Institute of Industrial Engineers
• /
• v.20 no.4
• /
• pp.5-21
• /
• 1994

We have described the optimal process route selection model for the PCB(printed circuit board) auto-insertion line. This PCB assembly line is known as a FFL(flexible flow line) which produces a range of products keeping the flow shop properties. Under FFL environments, we have emphasized the balancing of work-loads in order to maximize total productivity of PCB auto-insertion line. So we have developed a heuristic algorithm based on a work-order selection rule and min-max concept for the job route selection model.

• The Journal of Fisheries Business Administration
• /
• v.38 no.1 s.73
• /
• pp.19-45
• /
• 2007

There have not been many studies which considered both quantitative and qualitative location factors on the issues of site selection problems for a fish farm. This study develops AHP(analytic hierarchy process) model to resolve site selection problem for a fish raising farm by using quantitative and qualitative factors. In order to evaluate the validity of the location factors found in the literature review, the study used advice from fish raising farmers and related academic experts. Four major factors have been selected as economic factors, social factors, natural environmental factors and infrastructures. An AHP structural diagram has developed by considering the factors and potential sites proposed for fish farming. Through the survey on the preference of factors and potential sites, pairwise comparison matrices have been estimated and used to calculated the relative weights of each potential site. The AHP model process shown in the study can be applied to resolve site selection problems for fish raising farmers.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2002.05a
• /
• pp.569-574
• /
• 2002

As consumer's desire becomes various, agility of mold manufacturing is most important factor for competence of manufacturer. In common works to use commercial CAM system to generate tool path, some decision making process is required to produce optimal result of CAM systems, The paper proposes a methodology for computer-assisted tool selection procedures for various cutting type, such as rough, semi-rough and finish cuts. The system provides assist-tool-items for machining of design model part of injection meld die by analyzing sliced CAD model of die cavity and core. Also, the generating NC-code of the tool size is used to calculate machining time. The system is developed with commercial CAM using API. This module will be used for optimization of tool selection and planning process.