• The Korean Journal of Applied Statistics
• /
• v.25 no.4
• /
• pp.605-612
• /
• 2012

Multivariate data are often generated with multiple outcomes in various fields. Multiple outcomes could be mixed as continuous and discrete. Because of their complexity, the data are often dealt with by separately applying regression analysis to each outcome even though they are associated the each other. This univariate approach results in the low efficiency of estimates for parameters. We study the efficiency gains of the multivariate approaches relative to the univariate approach with the mixed data that include continuous and binary outcomes. All approaches yield consistent estimates for parameters with complete data. By jointly estimating parameters using multivariate methods, it is generally possible to obtain more accurate estimates for parameters than by a univariate approach. The association between continuous and binary outcomes creates a gap in efficiency between multivariate and univariate approaches. We provide a guidance to analyze the mixed data.

• The Korean Journal of Applied Statistics
• /
• v.28 no.2
• /
• pp.353-360
• /
• 2015

Various statistical models have been proposed over the last decade for spatially correlated Gaussian outcomes. The spatial linear mixed model (SLMM), which incorporates a spatial effect as a random component to the linear model, is the one of the most widely used approaches in various application contexts. Employing link functions, SLMM can be naturally extended to spatial generalized linear mixed model for non-Gaussian outcomes (SGLMM). We review popular SGLMMs on non-Gaussian spatial outcomes and demonstrate their applications with available public data.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.21 no.3
• /
• pp.11-46
• /
• 1996

The objectives of this research paper are to identify the types of the firm's new product development strategy and their characteristics about business strategy, to examine the effect of each type on new product outcomes, and to explore the contingency variable influencing the relationship between these types and new product outcomes. The result of the research are summarized as follows : First, in terms of both the resource allocation for product innovativeness and technology acquisition method, this study suggests 9 types of the firm's new product development strategies- Type 1 (pursuing low innovative products/relying on external technology), Type 2 (pursuing low innovative products oriented/relying on internal technology), Type 3 (pursuing low innovative products/relying on mixed technology), Type 4 (pursuing high innovative products/relying on internal technology), Type 6(pursuing high innovative products /relying on mixed technology), Type 7 (balancing low and high innovative products/relying on external technology), Type 8 (balancing low and high innovative products/relying on internal technology), Type 9 (balancing low and high innovative products/relying on mixed technology). Second, these 9 types are deeply associated with the firm's business strategic variables such as product differentiation and market differentiation, and exhibit different level of both technical and commercial performance of new products. Finally, the effects of these types on new product outcomes are different according to industrial environment and firms' characteristics with respect to size and technological capability.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.29 no.1
• /
• pp.161-167
• /
• 2001

The purpose of this study is to find out the difference of Zoysia matrella's evaporation in between 100 percent soil and mixed soil with 50 percent of perlite to create green spaces on the artificial ground. It is believed that the weight against the artificial ground will be reduced, provided the vegetation is possible in the circumstance of the mixed sol with 50 percent of perlite. The study employed a modified Makkink's model by Iwasa who had developed the model for estimating Zoysia matrella's evaporation in the natural ground using the Makkink's formula in 1997 at Chiba University, Japan. The parameter of Makkink's formula is the solar radiation. For that reason, the Makkink's formula is simple and easy to measure the parameter and has a high utility. If the outcomes from mixed soil are close to modified Makkinks formula, the modified Makkink's formula will be applied to estimate in the artificial ground with mixed soil with 50 percent of perlite. Weather observation and actual amount of evaporation of Zoysia matrella have been measured, and the relation between weather condition and actual amount of evaporation had been also investigated. In line with this, we found out that there is a relevant relationship between daily average temperature, the modified Makkink's model by Iwasa, and the actual amount of evaporation. As the results of the experiment, the outcomes from mixed soil with 50 percent of perlite have very high relation to 100 percent soil. In addition, mixed soil has more adhesion with water than natural soil. However, it needs to be adequately maintained in terms of fertilization and damage from disease and harmful insects until the gras fastens its roots into the soil. By using mixed soil with 50 percent of perlite, the load from soil on the artificial ground can be reduced. The study on the growth of the grass throughout the plant vegetation and the actual amount of evaporation in the mixed soil with 50 percent of perlite should be performed in the future.

• Journal of the Korean Data and Information Science Society
• /
• v.19 no.4
• /
• pp.1327-1334
• /
• 2008

Mixed effect binomial regression models are widely used for analysis of correlated count data in which the response is the result of a series of one of two possible disjoint outcomes. In this paper, we consider kernel extensions with nonparametric fixed effects and parametric random effects. The estimation is through the penalized likelihood method based on kernel trick, and our focus is on the efficient computation and the effective hyperparameter selection. For the selection of hyperparameters, cross-validation techniques are employed. Examples illustrating usage and features of the proposed method are provided.

• Journal of Korean Neurosurgical Society
• /
• v.61 no.1
• /
• pp.97-104
• /
• 2018

Objective : Diffusion-weighted magnetic resonance imaging (DW-MRI) has proven useful in the study of the natural history of ischemic stroke. However, the potential of DW-MRI for the evaluation of chronic subdural hematoma (CSDH) has not been established. In this study, we investigated DW-MRI findings of CSDH and evaluated the impact of the image findings on postoperative outcomes of CSDH. Methods : We studied 131 CSDH patients who had undergone single burr hole drainage surgery. The images of the subdural hematomas on preoperative DW-MRI and computed tomography (CT) were divided into three groups based on their signal intensity and density : 1) homogeneous (iso or low) density on CT and homogeneous low signal intensity on DW-MRI; 2) homogeneous (iso or low) density on CT and mixed signal intensity on DW-MRI; and 3) heterogeneous density on CT and mixed signal intensity on DW-MRI. On the basis of postoperative CT, we also divided the patients into 3 groups of surgical outcomes according to residual hematoma and mass effect. Results : Analysis showed statistically significant differences in surgical (A to B : p<0.001, A to C : p<0.001, B to C : p=0.129) and functional (A to B : p=0.039, A to C : p<0.001, B to C : p=0.108) outcomes and treatment failure rates (A to B : p=0.037, A to C : p=0.03, B to C : p=1) between the study groups. In particular, group B and group C showed worse outcomes and higher treatment failure rates than group A. Conclusion : CSDH with homogeneous density on CT was characterized by signal intensity on DW-MRI. In CSDH patients, performing DW-MRI as well as CT helps to predict postoperative treatment failure or complications.

• Communications for Statistical Applications and Methods
• /
• v.19 no.6
• /
• pp.761-770
• /
• 2012

We frequently encounter outcomes of count that have extra variation. This paper considers several alternative models for overdispersed count responses such as a quasi-Poisson model, zero-inflated Poisson model and a negative binomial model with a special focus on a generalized linear mixed model. We also explain various goodness-of-fit criteria by discussing their appropriateness of applicability and cautions on misuses according to the patterns of response categories. The overdispersion models for counts data have been explained through two examples with different response patterns.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.4
• /
• pp.927-936
• /
• 2017

To analyze longitudinal count data, Poisson linear mixed models are commonly used. In the models the random effects covariance matrix explains both within-subject variation and serial correlation of repeated count outcomes. When the random effects covariance matrix is assumed to be misspecified, the estimates of covariates effects can be biased. Therefore, we propose reasonable and flexible structures of the covariance matrix using autoregressive and moving average Cholesky decomposition (ARMACD). The ARMACD factors the covariance matrix into generalized autoregressive parameters (GARPs), generalized moving average parameters (GMAPs) and innovation variances (IVs). Positive IVs guarantee the positive-definiteness of the covariance matrix. In this paper, we use the ARMACD to model the random effects covariance matrix in Poisson loglinear mixed models. We analyze epileptic seizure data using our proposed model.

• Journal of Family Resource Management and Policy Review
• /
• v.26 no.2
• /
• pp.19-36
• /
• 2022

The purpose of this study was (a) to identify latent profiles among employed Korean mothers of third graders based on work-family conflict, work-family enrichment, work-parenting conflict, and work-parenting enrichment, (b) to examine the antecedents of profile membership at the individual, family, work, and community levels, and (c) to investigate the differences in the various psychological outcomes across the profiles. The sample of 451 married employed mothers was a subset of data from the 10th Wave of the Panel Study of Korean Children, which was collected in 2017 when the focal child was in the third grade. Our latent profile analysis suggested a three-profile model that comprised enriched (11.91%), moderate (47.85%), and mixed (40.24%) profiles. The significant antecedents of profile membership were subjective health status, the child's adjustment to school, working hours, the community's suitability for childrearing, and satisfaction with community service facilities. Regarding psychological outcomes, the levels of life satisfaction, marital satisfaction, and job satisfaction were higher in the following order: enriched, moderate, and mixed profiles. The levels of depressive symptoms were in the reverse order: mixed, moderate, and enriched profiles. This study contributes to a comprehensive understanding of the literature on work-family interactions by considering various predictors and outcomes at multiple levels.

• International Journal of Highway Engineering
• /
• v.14 no.5
• /
• pp.145-155
• /
• 2012

PURPOSES: This study proposes a novel method based on microscopic simulation models to evaluate bicycle passing ways in mixed traffic flow conditions at signalized intersections. METHODS: Both operational efficiency and safety are taken into consideration in the evaluation. A widely used performance measure, delay, is used for evaluating the operational efficiency. Regarding the safety evaluation, surrogate safety measures (SSM) to represent traffic conflicts and the level of crash severity, DeltaS and Max.DeltaV, are applied in the proposed method. RESULTS: Extensive simulations and statistical tests show that an integrated bike-box way is identified as the best in terms of operational efficiency and safety. CONCLUSIONS: The proposed method and outcomes of this study will be valuable for bicycle traffic operations and facility design.