• Journal of the Korean Data and Information Science Society
• /
• v.25 no.6
• /
• pp.1263-1272
• /
• 2014

Longitudinal categorical data commonly occur from medical, health, and social sciences. In these data, the correlation of repeated outcomes is taken into account to explain the effects of covariates exactly. In this paper, we introduce marginalized random effects models that are used for the estimation of the population-averaged effects of covariates. We also review how these models have been developed. Real data analysis is presented using the marginalized random effects.

• The Korean Journal of Applied Statistics
• /
• v.35 no.6
• /
• pp.703-724
• /
• 2022

Although longitudinal studies mainly produce multivariate longitudinal data, most of existing statistical models analyze univariate longitudinal data and there is a limitation to explain complex correlations properly. Therefore, this paper describes various methods of modeling the covariance matrix to explain the complex correlations. Among them, modified Cholesky decomposition, modified Cholesky block decomposition, and hypersphere decomposition are reviewed. In this paper, we review these methods and analyze Korean children and youth panel (KCYP) data are analyzed using the Bayesian method. The KCYP data are multivariate longitudinal data that have response variables: School adaptation, academic achievement, and dependence on mobile phones. Assuming that the correlation structure and the innovation standard deviation structure are different, several models are compared. For the most suitable model, all explanatory variables are significant for school adaptation, and academic achievement and only household income appears as insignificant variables when cell phone dependence is a response variable.

• Genomics & Informatics
• /
• v.20 no.1
• /
• pp.8.1-8.14
• /
• 2022

Despite the success of recent genome-wide association studies investigating longitudinal traits, a large fraction of overall heritability remains unexplained. This suggests that some of the missing heritability may be accounted for by gene-gene and gene-time/environment interactions. In this paper, we develop a Bayesian variable selection method for longitudinal genetic data based on mixed models. The method jointly models the main effects and interactions of all candidate genetic variants and non-genetic factors and has higher statistical power than previous approaches. To account for the within-subject dependence structure, we propose a grid-based approach that models only one fixed-dimensional covariance matrix, which is thus applicable to data where subjects have different numbers of time points. We provide the theoretical basis of our Bayesian method and then illustrate its performance using data from the 1000 Genome Project with various simulation settings. Several simulation studies show that our multivariate method increases the statistical power compared to the corresponding univariate method and can detect gene-time/ environment interactions well. We further evaluate our method with different numbers of individuals, variants, and causal variants, as well as different trait-heritability, and conclude that our method performs reasonably well with various simulation settings.

• Journal of Internet Computing and Services
• /
• v.21 no.4
• /
• pp.77-85
• /
• 2020

The purpose of this study is to explore the changing trajectory of using educational media through longitudinal data analysis. We categorize the feature of using educational media as usage for learning, usage for information, and usage for the game. We explore the longitudinal changing patterns of usage for learning, usage for information, and usage for the game by LGM(Longitudinal Growth Modeling). We also find the gender difference between these longitudinal changing trajectories. We used 3,499 samples of KYPS middle school second-grade panel data. We found these results: (a) Both usage for learning and information are statically significant variability in initial level and rate of change. Both of the changing trajectories have increased. (b) Girls have a higher rate of the change both in the usage of learning and information than boys over time. (c) There is a statistically significant individual variability in initial levels and rate of change in the usage of the game over time. (d) Boys have a higher rate of initial value than girls in the usage of games, but there is no significant difference in the rate of changing trajectories.

• Journal of the Korea Convergence Society
• /
• v.10 no.8
• /
• pp.137-144
• /
• 2019

This paper deals with the longitudinal study on the supply & demand-side diversity of digital media service. The purpose of this study is to measure the diversity of contents supply and demand-side of digital media platform providers by longitudinal data and discuss the implication. In the approval and re-authorization of pay-TV broadcasters, there were attempts to measure diversity indicators as items to evaluate the publicness and public interest of broadcasting, but they are mainly limited to the method of measuring diversity in the short-term supply side. Thus researcher wants to confirm the evaluation of two aspects through this study. First, researcher proposes a demand-side measurement methodology that utilizes actual audience data from users, and second, a longitudinal evaluation methodology that evaluates long-term trends of diversity change. Researcher has secured the actual supply and demand data of the platform player and confirmed trends of longitudinal diversification indexing for 50 months from 2012 to 2017. Through this research, researcher expects that the supply and demand-side and the longitudinal diversity evaluation will be utilized in a balanced way of publicness and public interest evaluation of broadcasting.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.26 no.1
• /
• pp.109-116
• /
• 2001

The detection of corporate failures is a subject that has been particularly amenable to cross-sectional financial ratio analysis. In most of firms, however, the financial data are available over past years. Because of this, a model utilizing these longitudinal data could provide useful information on the prediction of bankruptcy. To correctly reflect the longitudinal and firm-specific data, the generalized linear model with assuming the first order AR(autoregressive) process is proposed. The method is motivated by the clinical research that several characteristics are measured repeatedly from individual over the time. The model is compared with several other predictive models to evaluate the performance. By using the financial data from manufacturing corporations in the Korea Stock Exchange (KSE) list, we will discuss some experiences learned from the procedure of sampling scheme, variable transformation, imputation, variable selection, and model evaluation. Finally, implications of the model with repeated measurement and future direction of research will be discussed.

• Journal of the Korean Statistical Society
• /
• v.26 no.3
• /
• pp.319-332
• /
• 1997

The application of multivariate linear rank statistics to data with item nonresponse is considered. Only a modest extension of the complete data techniques is required when the missing data may be thought of as a random sample, and an appropriate modification of the covariances is derived. A proof of the asymptotic multivariate normality is given. A review of some related results in the literature is presented and applications including longitudinal and repeated measures designs are discussed.

• The Korean Journal of Applied Statistics
• /
• v.28 no.2
• /
• pp.335-342
• /
• 2015

Joint hierarchical generalized linear models proposed by Molas et al. (2013) extend the simple longitudinal model into multiple models fitted jointly. It can easily handle the correlation of multivariate longitudinal data. In this paper, we apply this method to analyze KoGES cohort dataset. Fixed unknown parameters, random effects and variance components are estimated based on a standard framework of h-likelihood theory. Furthermore, based on the conditional Akaike information criterion the correlated covariance structure of random-effect model is selected rather than an independent structure.

• International Journal of Computer Science & Network Security
• /
• v.21 no.4
• /
• pp.52-58
• /
• 2021

Questionnaires and surveys are increasingly being used to collect information from participants of empirical software engineering studies. Usually, such data is analyzed using statistical methods to show an overall picture of participants' agreement or disagreement. In general, the whole survey population is considered as one group with some methods to extract varieties. Sometimes, there are different opinions in the same group, but they are not well discovered. In some cases of the analysis, the population may be divided into subgroups according to some data. The opinions of different segments of the population may be the same. Even though the existing approach can capture the general trends, there is a risk that the opinions of different sub-groups are lost. The problem becomes more complex in longitudinal studies where minority opinions might fade over time. Longitudinal survey data may include several interesting patterns that can be extracted using a clustering process. It can discover new information and give attention to different opinions. We suggest using a data mining approach to finding the diversity among the different groups in longitudinal studies. Our study shows that diversity can be revealed and tracked over time using the clustering approach, and the minorities have an opportunity to be heard.

• Korean Journal of Child Studies
• /
• v.37 no.6
• /
• pp.5-17
• /
• 2016

Objective: This study focused on the longitudinal associations between self-esteem and peer relationships in Korean adolescents while considering gender and timing-early and late adolescence-differences. Methods: The study made use of data from the Korean Children and Youth Panel Survey. Three waves of data collected from 2,351 adolescents were analyzed by means of autoregressive cross-lagged modeling. Results: The results indicated that self-esteem predicted subsequent changes in peer relationship but not vice versa. Further, the results that longitudinal associations between self-esteem and peer relationships differed between male and female adolescents and between early and late adolescence. Conclusion: The findings revaluated the longitudinal relationship between self-esteem and peer relationships. Both gender and timing should be considered when planning interventions related issues about self-esteem and peer relationships in adolescence.