• Communications for Statistical Applications and Methods
• /
• 제10권1호
• /
• pp.125-134
• /
• 2003

Standard estimation methods for linear mixed models are sensitive to influential observations. However, tools and concepts for linear mixed model diagnostics are rudimentary until now and research is heavily demanded in linear mixed models. In this paper, we consider two diagnostics to evaluate the effects of individual observations in the estimation of fixed effects for linear mixed models. Those are Cook's distance and COVRATIO. Results of our limited simulation study suggest that the Cook's distance is not good statistical quantity in linear mixed models. Also calibration point for COVRATIO seems to be quite conservative.

• Journal of the Korean Data and Information Science Society
• /
• 제28권4호
• /
• 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.

• Communications for Statistical Applications and Methods
• /
• 제15권6호
• /
• pp.969-976
• /
• 2008

Mixed linear models have been widely used in various correlated data including multivariate survival data. In this paper we extend hierarchical-likelihood(h-likelihood) approach for mixed linear models with right censored data to that for left censored data. We also allow a general random-effect structure and propose the estimation procedure. The proposed method is illustrated using a numerical data set and is also compared with marginal likelihood method.

• Journal of the Korean Statistical Society
• /
• 제28권2호
• /
• pp.211-223
• /
• 1999

We propose a simple estimation procedure in the mixed linear models with censored normal data, using both Buckly and James(1979) type pseudo random variables and Lee and Nelder's(1996) estimation procedure. The proposed method is illustrated with the matched pairs data in Pettitt(1986).

• 응용통계연구
• /
• 제28권2호
• /
• pp.123-136
• /
• 2015

관측 가능한 변수들 사이의 관계를 묘사한 갈릴레오의 물리학 법칙 발견 이후, 과학은 큰 성과를 거두며 발전해왔다. 그러나, 관측할 수 없는 변량효과를 함께 이용하여 더 많은 자연 현상을 설명할 수 있게 되었고, 이를 이용한 최초의 통계적 모형인 혼합효과모형이 소개되었다. 계산기술의 발달과 더불어 복잡한 현상에 대한 추론을 위하여 혼합효과모형은 그 중요성이 더욱 커지고 있다. 이러한 혼합효과모형은 최근 다단계 일반화 선형모형을 포함한 여러 모형으로 확장되었으며, 관측할 수 없는 변량효과를 추론하기 위한 다단계 가능도가 제시되었다. 혼합효과모형 특집호를 통해 이러한 모형들이 여러 통계학적 문제점을 해결하는 과정을 제시하고, 앞으로 어떤 확장이 추가적으로 요구되는 지에 대하여 논할 것이다. 빈도록적 접근법과 베이지안 접근법을 함께 다룬다.

• Communications for Statistical Applications and Methods
• /
• 제24권1호
• /
• pp.81-96
• /
• 2017

Cumulative logit random effects models are typically used to analyze longitudinal ordinal data. The random effects covariance matrix is used in the models to demonstrate both subject-specific and time variations. The covariance matrix may also be homogeneous; however, the structure of the covariance matrix is assumed to be homoscedastic and restricted because the matrix is high-dimensional and should be positive definite. To satisfy these restrictions two Cholesky decomposition methods were proposed in linear (mixed) models for the random effects precision matrix and the random effects covariance matrix, respectively: modified Cholesky and moving average Cholesky decompositions. In this paper, we use these two methods to model the random effects precision matrix and the random effects covariance matrix in cumulative logit random effects models for longitudinal ordinal data. The methods are illustrated by a lung cancer data set.

• Journal of the Korean Data and Information Science Society
• /
• 제12권1호
• /
• pp.19-25
• /
• 2001

Random effects models which describe the dependence via random effects in various correlated data have recently received considerable attention in the biomedical literature. They include mixed linear models (MLMs), generatized linear mixed models (GLMMS) and hierarchical generalized linear models (HGLMs). For the inference Lee and Nelder (2000) proposed the first-and second-order REML (restricted maximum likelihood) methods based on hierarchical-likelihood of tee and Welder (1996). In this paper, for Poisson-gamma HGLMs the new methods are theoretically compared with marginal likelihood methods and both methods are illustrated by two practical examples.

• Journal of the Korean Data and Information Science Society
• /
• 제20권2호
• /
• pp.445-452
• /
• 2009

Credibility models are actuarial tools to distribute premiums fairly among a heterogeneous group of policyholders. Many existing credibility models can be expressed as special cases of linear mixed effects models. In this paper we propose a nonlinear credibility regression model by reforming the linear mixed effects model through kernel machine. The proposed model can be seen as prediction method applicable in any setting where repeated measures are made for subjects with different risk levels. Experimental results are then presented which indicate the performance of the proposed estimating procedure.

• Communications for Statistical Applications and Methods
• /
• 제22권5호
• /
• pp.507-518
• /
• 2015

It is not easy to select a linear mixed model since the main interest for model building could be different and the number of parameters in the model could not be clearly defined. In this paper, performance of conditional Akaike Information Criteria and its bias-corrected version are compared with marginal Bayesian and Akaike Information Criteria through a simulation study. The results from the simulation study indicate that bias-corrected conditional Akaike Information Criteria shows promising performance when candidate models exclude large models containing the true model, but bias-corrected one prefers over-parametrized models more intensively when a set of candidate models increases. Marginal Bayesian and Akaike Information Criteria also have some difficulty to select the true model when the design for random effects is nested.

• Communications for Statistical Applications and Methods
• /
• 제21권5호
• /
• pp.423-433
• /
• 2014

Test statistics using cumulative sums of residuals have been widely used in various regression models including generalized linear models(GLM). Recently, Pan and Lin (2005) extended this testing procedure to the generalized linear mixed models(GLMM) having random effects, in which we encounter difficulties in computing the marginal likelihood that is expressed as an integral of random effects distribution. The Gaussian quadrature algorithm is commonly used to approximate the marginal likelihood. Many commercial statistical packages provide an option to apply this type of goodness-of-fit test in GLMs but available programs are very rare for GLMMs. We suggest a computational algorithm to implement the testing procedure in GLMMs by a freely accessible R package, and also illustrate through practical examples.