• The Korean Journal of Applied Statistics
• /
• v.28 no.2
• /
• pp.211-219
• /
• 2015

Generalized linear mixed models are used to analyze longitudinal categorical data. Random effects specify the serial dependence of repeated outcomes in these models; however, the estimation of a random effects covariance matrix is challenging because of many parameters in the matrix and the estimated covariance matrix should satisfy positive definiteness. Several approaches to model the random effects covariance matrix are proposed to overcome these restrictions: modified Cholesky decomposition, moving average Cholesky decomposition, and partial autocorrelation approaches. We review several approaches and present potential future work.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.1
• /
• pp.1-10
• /
• 2017

Longitudinal data of Korean child academic achievement have been used to find the significant exploratory variables under the assumption of independent repeated measured data. Using the exploratory variables in previous research works, we analyze the linear mixed model incorporating the fixed and random effects for child academic achievement to detect the significant exploratory variables. Korea welfare panel study data observed three times between 2006 and 2012 by additional survey for children. The child academic achievement is evaluated by the sum of academic achievements of Korean, English and Mathematics. We also investigate the multicollinearity and the missing mechanism and select some popular correlation matrices to analyze the linear mixed model.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.5
• /
• pp.1055-1067
• /
• 2017

Most previous works to study for the depression of the disabilities in Korea have analyzed the repeated measured data of each individual under the mutually independent assumption. In this study, Korea Welfare Panel data of the disabilities surveyed additionally every three years are analyzed to detect the significant exploratory variables by the linear mixed models. A suitable correlation matrix is considered for the dependency of repeated measurement of each individual. The random effect to reflect the characteristics of the individuals as well as the fixed effect is included in the fitted linear mixed model. By the residual plot of the fixed effect model, the problem that the averages of residuals of each individual do not seem to be around zero is described. Further, the residual plot and the Q-Q plot coming from the selected final model are shown that the problem is modified well.

• Management & Information Systems Review
• /
• v.29 no.4
• /
• pp.207-223
• /
• 2010

One of the important prudential regulations is the capital regulation. The current domestic and international capital regulation sets the minimum capital requirement according to the size of risk which is the simple sum of market risk and credit risk. However the portfolio theory suggests that, due to the effect of diversification, the total risk is less than the summation of market and credit risk. This paper investigates and does empirical test to verify the diversification effect in measuring financial firm's integrated risk. We verify the diversification effect between the market risk and credit risk. This paper's contribution is to present the empirical evidence that, considering the relationship between market and credit risk, the integrated risk is less than sum of them. This implication is that the surplus capital may be used for the other purposes, therefore enhancing capital allocation efficiency in view of society as a whole.

• Journal of Labour Economics
• /
• v.43 no.4
• /
• pp.179-203
• /
• 2020

This study aims to compare and analyze the impact of labor unions on job training programs using two different longitudinal data, Workplace Panel Survey (WPS) and Korean Labor and Inocme Panel Study (KLIPS). By applying the Correlated Random Effect (CRE) model to both individual-level and establishment-level data, we observe that labor unions increase the likelihood that establishments provide employees with job training programs and workers participate in the trainings. Our results shed light on the role of labor unions to increase the opportunities of job training programs for workers.

• The Korean Journal of Applied Statistics
• /
• v.21 no.6
• /
• pp.923-932
• /
• 2008

We are likely to face complex multivariate data which can be characterized by having a non-trivial correlation structure. For instance, omitted covariates may simultaneously affect more than one count in clustered data; hence, the modeling of the correlation structure is important for the efficiency of the estimator and the computation of correct standard errors, i.e., valid inference. A standard way to insert dependence among counts is to assume that they share some common unobservable variables. For this assumption, we fitted correlated random effect models considering multilevel model. Estimation was carried out by adopting the semiparametric approach through a finite mixture EM algorithm without parametric assumptions upon the random coefficients distribution.

• Journal of Animal Science and Technology
• /
• v.50 no.2
• /
• pp.151-156
• /
• 2008

The study aimed to estimate genetic parameters of body weights in Hanwoo steers using random regression model and compare it with single trait animal model. A total of 1,372 Hanwoo steers that belonged to progeny testing program of the Hanwoo Genetic Improvement conducted at the Livestock Improvement Main Center of the National Agricultural Cooperative Federation (LIMC-NACF) in Rep. of Korea were used. Results of the random regression model fitting quadratic function revealed heritability values from 0.17 to 0.30 for the whole testing days up to 800 days. The results of the animal model showed estimated heritability values ranged from 0.24 to 0.36. Estimates of permanent environmental correlations tended to increase with increasing test in days. Unlike in the direct genetic correlation that at early stage the estimate was slightly negative it was 0.30 then increased to approach unity at later stage of test. Comparing the results between random regression model and the animal model showed not much differences and both followed similar pattern and therefore the use of random regression model for the national genetic evaluation of Hanwoo could be implemented.

• The Korean Journal of Applied Statistics
• /
• v.22 no.1
• /
• pp.181-188
• /
• 2009

This paper discusses the covariance structures of data collected from an experiment with a nested design structure, where a smaller experimental unit is nested within a larger one. Due to the nonrandomization of repeated measures factors to the nested experimental units, compound symmetry covariance structure is assumed for the analysis of data. Treatments are given as the combinations of the levels of random factors and fixed factors. So, a mixed-effects model is suggested under compound symmetry structure. An example is presented to illustrate the nesting in the experimental units and to show how to get the parameter estimates in the fitted model.

• Economic Analysis
• /
• v.27 no.1
• /
• pp.37-66
• /
• 2021

This paper examines the potential impacts of COVID-19 on poverty in ASEAN countries. The first estimate, adopted from Summer et al. (2020) and Nonvide (2020), configures three scenarios of contractions in per capita household income or consumption; the impact of each scenario on poverty is calculated using poverty lines at different thresholds. In the second estimate, poverty impacts in 2020 and 2021 were projected using regression models controlling for unobserved country effects, unbalanced data, and endogeneity. COVID-19 has been shown to have negative impacts on poverty reduction in the ASEAN Member States. To reduce poverty, concerted efforts are needed to implement policies for reducing income inequality and promoting economic growth. Such efforts will not only speed up the countries' return to pre-pandemic poverty levels but also contribute to further accelerating poverty reduction.

• Journal of Animal Science and Technology
• /
• v.49 no.2
• /
• pp.177-182
• /
• 2007

The objective of this study was to evaluate the genetic relationship between weaning weight (WW) and carcass traits. Carcass traits were eye muscle area(EMA), back fat thickness(BFT), marbling score 1(MS1) in 21 grade scales, marbling score 2(MS2) in 7 grade scales and meat color scores(Mcolor). Parameters were estimated by REML procedure with MTDFREML package. Models included contemporary group as defined by the same year-season-sex at birth, linear covariates of age(days) at weaning, age of dam(days) and age at slaughter(days) as fixed effects and animal random effects for all the traits. Heritability estimates of WW, EMA, BFT, MS1, MS2 and Mcolor were 0.25, 0.20, 0.20, 0.32, 0.32 and 0.22, respectively. Genetic(phenotypic) correlation coefficients of WW with EMA, BFT, MS1, MS2 and Mcolor were 0.75(0.16), 0.18(0.05), －0.41(－0.09), －0.40(0.11) and －0.07(0.05), respectively. Results from this study suggest that single trait selection for weaning weight would lead to progeny population having carcass with large EMA, thick BFT and decreased MS.