• Communications for Statistical Applications and Methods
• /
• v.27 no.5
• /
• pp.523-533
• /
• 2020

This paper suggests three available methods for finding nonnegative estimates of variance components of the random effects in mixed models. The three proposed methods based on the concepts of projections are called projection method I, II, and III. Each method derives sums of squares uniquely based on its own method of projections. All the sums of squares in quadratic forms are calculated as the squared lengths of projections of an observation vector; therefore, there is discussion on the decomposition of the observation vector into the sum of orthogonal projections for establishing a projection model. The projection model in matrix form is constructed by ascertaining the orthogonal projections defined on vector subspaces. Nonnegative estimates are then obtained by the projection model where all the coefficient matrices of the effects in the model are orthogonal to each other. Each method provides its own system of linear equations in a different way for the estimation of variance components; however, the estimates are given as the same regardless of the methods, whichever is used. Hartley's synthesis is used as a method for finding the coefficients of variance components.

• Communications for Statistical Applications and Methods
• /
• v.23 no.6
• /
• pp.575-585
• /
• 2016

Longitudinal studies repeatedly measure outcomes over time. Therefore, repeated measurements are serially correlated from same subject (within-subject variation) and there is also variation between subjects (between-subject variation). The serial correlation and the between-subject variation must be taken into account to make proper inference on covariate effects (Diggle et al., 2002). However, estimation of the covariance matrix is challenging because of many parameters and positive definiteness of the matrix. To overcome these limitations, we propose autoregressive moving average Cholesky decomposition (ARMACD) for the linear mixed models. The ARMACD allows a class of flexible, nonstationary, and heteroscedastic models that exploits the structure allowed by combining the AR and MA modeling of the random effects covariance matrix. We analyze a real dataset to illustrate our proposed methods.

• Communications for Statistical Applications and Methods
• /
• v.12 no.3
• /
• pp.759-771
• /
• 2005

Twin studies are one of the most widely used methods for quantifying the influence of genetic and environmental factors on some traits such as a life span or a disease. In this paper we propose a genetic mixed linear model for twin survival time data, which allows us to separate the genetic component from the environmental component. Inferences are based upon the hierarchical likelihood (h-likelihood), which provides a statistically efficient and simple unified framework for various random-effect models. We also propose a simple and fast computation method for analyzing a large data set on twin survival study. The new method is illustrated to the survival data in Swedish Twin Registry. A simulation study is carried out to evaluate the performance.

• The Korean Journal of Applied Statistics
• /
• v.27 no.6
• /
• pp.923-932
• /
• 2014

The Hurdle model can to analyze zero-inflated count data. This model is a mixed model of the logit model for a binary component and a truncated Poisson model of a truncated count component. We propose a new hurdle model with a general heterogeneous random effects covariance matrix to analyze longitudinal zero-inflated count data using modified Cholesky decomposition. This decomposition factors the random effects covariance matrix into generalized autoregressive parameters and innovation variance. The parameters are modeled using (generalized) linear models and estimated with a Bayesian method. We use these methods to carefully analyze a real dataset.

• Communications for Statistical Applications and Methods
• /
• v.30 no.1
• /
• pp.95-107
• /
• 2023

Purpose: The statistical analysis of point processes on linear networks is a recent area of research that studies processes of events happening randomly in space (or space-time) but with locations limited to reside on a linear network. For example, traffic accidents happen at random places that are limited to lying on a network of streets. This paper applies techniques developed for point processes on linear networks and the tools available in the R-package spatstat to estimate the intensity of traffic accidents in Leon County, Florida. Methods: The intensity of accidents on the linear network of streets is estimated using log-linear Poisson models which incorporate cubic basis spline (B-spline) terms which are functions of the x and y coordinates. The splines used equally-spaced knots. Ten different models are fit to the data using a variety of covariates. The models are compared with each other using an analysis of deviance for nested models. Results: We found all covariates contributed significantly to the model. AIC and BIC were used to select 9 as the number of knots. Additionally, covariates have different effects such as increasing the speed limit would decrease traffic accident intensity by 0.9794 but increasing the number of lanes would result in an increase in the intensity of traffic accidents by 1.086. Conclusion: Our analysis shows that if other conditions are held fixed, the number of accidents actually decreases on roads with higher speed limits. The software we currently use allows our models to contain only spatial covariates and does not permit the use of temporal or space-time covariates. We would like to extend our models to include such covariates which would allow us to include weather conditions or the presence of special events (football games or concerts) as covariates.

• 한국데이터정보과학회:학술대회논문집
• /
• 2006.04a
• /
• pp.37-37
• /
• 2006

• Asian-Australasian Journal of Animal Sciences
• /
• v.13 no.3
• /
• pp.413-422
• /
• 2000

In the class of models which include random effects, the variance component estimates are important to obtain accurate predictors and estimators. Variance component estimation is straightforward for balanced data but not for unbalanced data. Since orthogonality among factors is absent in unbalanced data, various methods for variance component estimation are available. REML estimation is the most widely used method in animal breeding because of its attractive statistical properties. Recently, Bayesian approach became feasible through Markov Chain Monte Carlo methods with increasingly powerful computers. Furthermore, advances in variance component estimation with complicated models such as generalized linear mixed models enabled animal breeders to analyze non-normal data.

• Journal of the Korean Data and Information Science Society
• /
• v.25 no.4
• /
• pp.941-949
• /
• 2014

Unit level logistic regression model with mixed effects has been used for estimating small area proportions, which treats the spatial effects as random effects and assumes linearity between the logistic link and the covariates. However, when the functional form of the relationship between the logistic link and the covariates is not linear, it may lead to biased estimators of the small area proportions. In this paper, we relax the linearity assumption and propose two types of kernel-based logistic regression models for estimating small area proportions. We also demonstrate the efficiency of our propose models using simulated data and real data.

• The Journal of Fisheries Business Administration
• /
• v.51 no.1
• /
• pp.81-96
• /
• 2020

The purpose of this paper is to analyze factors affecting the net income of regional and industrial fisheries cooperatives in South Korea using panel data. This paper utilizes linear or GLS regression models such as pooled OLS model, fixed effects model, and random effects model to estimate affecting factors of the net income of regional and industrial fisheries cooperatives. After reviewing various tests, we eventually select random effects model. The results, based on panel data between 2013 and 2018 year and 64 fisheries cooperatives, indicate that capital and area dummy variables have positive effects and employment has negative effect on the net income of regional and industrial fisheries cooperatives as predicted. However, debt are opposite with our predictions. Specifically, it turns out that debt has positive effect on the net income of regional and industrial fisheries cooperatives although it has been increased. Additionally, this paper shows that the member of confreres does not show any significant effect on the net income of regional and industrial fisheries cooperatives in South Korea. This study is significant in that it analyzes the major factors influencing changes in the net income that have not been conducted recently for the fisheries cooperatives by region and industry.

• Asian-Australasian Journal of Animal Sciences
• /
• v.21 no.9
• /
• pp.1227-1232
• /
• 2008

Carcass characteristics data of Hanwoo (N = 1,084) were collected from two stations of the National Livestock Institute of Animal Science (NIAS), Korea and records from thirteen individual cow-calf operators were analyzed to estimate variance and covariance components and the effect of different slaughter endpoints. Carcass traits analyzed were cold carcass weight (CWT, kg), REA (rib eye area, cm2), back fat thickness (mm) and marbling score (1-7). Four different models were examined. All models included sex and contemporary group as fixed effects and the animal's direct genetic potential and environment as random effects. The first model fitted a linear covariate of age at slaughter. The second model fitted both linear and quadratic covariates of age at slaughter. The third model fitted a linear covariate of body weight at slaughter. The fourth model fitted both linear covariates of age at slaughter and body weight at slaughter. Variance components were estimated using the REML procedure with Gibb's sampler. Heritability estimate of CWT was in the range of 0.08-0.11 depending on the model applied. Heritability estimates of BF, REA and MS were in the ranges of 0.23-0.28, 0.19-0.26, and 0.44-0.45, respectively. Genetic correlations between CWT and BF, between CWT and REA, and between CWT and MS were in the ranges of -0.33 - -0.14, 0.73-0.84, and -0.01- 0.11, respectively. Genetic correlations between REA and BF, between MS and BF and between REA and MS were in the ranges of -0.82 ~ -0.72, 0.04~0.28 and -0.08 ~ -0.02, respectively. Variance and covariance components estimated varied by model with different slaughter endpoints. Body weight endpoint was more effective for direct selection in favor of yield traits and body weight endpoints affected more of the correlated response to selection for the traits of yield and quality of edible portion of beef.