• Journal of the Korean Data and Information Science Society
• /
• 제20권5호
• /
• pp.905-912
• /
• 2009

We propose a doubly penalized kernel machine (DPKM) which uses heteroscedastic location-scale model as basic model and estimates both mean and volatility functions simultaneously by kernel machines. We also present the model selection method which employs the generalized approximate cross validation techniques for choosing the hyperparameters which affect the performance of DPKM. Artificial examples are provided to indicate the usefulness of DPKM for the mean and volatility functions estimation.

• Journal of the Korean Data and Information Science Society
• /
• 제27권2호
• /
• pp.499-510
• /
• 2016

생존분석 회귀모형에서 적절한 변수를 선택하는 것은 매우 중요하다. 본 논문에서는 "frailtyHL" R 패키지 (Ha 등, 2012)를 기반으로 하여 다수준 프레일티 모형 (multi-level frailty models)에서 벌점화 변수선택 방법 (penalized variable-selection method)의 절차를 소개한다. 여기서 모형 추정은 벌점화 다단계 가능도에 기초하며, 세 가지 벌점 함수 (LASSO, SCAD 및 HL)가 고려된다. 개발된 방법의 예증을 위해 벨기에 EORTC (European Organization for Research and Treatment of Cancer; 유럽 암 치료기구)에서 수행된 다국가/다기관 임상시험 자료를 이용하여 세 가지 변수 선택 방법의 결과를 비교하고, 그 결과들의 상대적 장 단점에 대해 토론한다. 특히, 자료 분석 결과에 의하면 SCAD와 HL방법이 LASSO보다 중요한 변수를 잘 선택하는 것으로 나타났다.

• Genomics & Informatics
• /
• 제14권4호
• /
• pp.138-148
• /
• 2016

The success of genome-wide association studies (GWASs) has enabled us to improve risk assessment and provide novel genetic variants for diagnosis, prevention, and treatment. However, most variants discovered by GWASs have been reported to have very small effect sizes on complex human diseases, which has been a big hurdle in building risk prediction models. Recently, many statistical approaches based on penalized regression have been developed to solve the "large p and small n" problem. In this report, we evaluated the performance of several statistical methods for predicting a binary trait: stepwise logistic regression (SLR), least absolute shrinkage and selection operator (LASSO), and Elastic-Net (EN). We first built a prediction model by combining variable selection and prediction methods for type 2 diabetes using Affymetrix Genome-Wide Human SNP Array 5.0 from the Korean Association Resource project. We assessed the risk prediction performance using area under the receiver operating characteristic curve (AUC) for the internal and external validation datasets. In the internal validation, SLR-LASSO and SLR-EN tended to yield more accurate predictions than other combinations. During the external validation, the SLR-SLR and SLR-EN combinations achieved the highest AUC of 0.726. We propose these combinations as a potentially powerful risk prediction model for type 2 diabetes.

• Genomics & Informatics
• /
• 제20권4호
• /
• pp.48.1-48.7
• /
• 2022

Penalized regression has been widely used in genome-wide association studies for joint analyses to find genetic associations. Among penalized regression models, the least absolute shrinkage and selection operator (Lasso) method effectively removes some coefficients from the model by shrinking them to zero. To handle group structures, such as genes and pathways, several modified Lasso penalties have been proposed, including group Lasso and sparse group Lasso. Group Lasso ensures sparsity at the level of pre-defined groups, eliminating unimportant groups. Sparse group Lasso performs group selection as in group Lasso, but also performs individual selection as in Lasso. While these sparse methods are useful in high-dimensional genetic studies, interpreting the results with many groups and coefficients is not straightforward. Lasso's results are often expressed as trace plots of regression coefficients. However, few studies have explored the systematic visualization of group information. In this study, we propose a multi-level polar Lasso (MP-Lasso) chart, which can effectively represent the results from group Lasso and sparse group Lasso analyses. An R package to draw MP-Lasso charts was developed. Through a real-world genetic data application, we demonstrated that our MP-Lasso chart package effectively visualizes the results of Lasso, group Lasso, and sparse group Lasso.

• 응용통계연구
• /
• 제35권2호
• /
• pp.217-227
• /
• 2022

분위수 회귀 모형은 변수에 숨겨진 복잡한 정보를 살펴보기 위한 효율적인 도구를 제공하는 장점을 바탕으로 많은 분야에서 널리 사용되고 있다. 그러나 현대의 대용량-고차원 데이터는 계산 시간 및 저장공간의 제한으로 인해 분위수 회귀 모형의 추정을 매우 어렵게 만든다. 분할-정복은 전체 데이터를 계산이 용이한 여러개의 부분집합으로 나눈 다음 각 분할에서의 요약 통계량만을 이용하여 전체 데이터의 추정량을 재구성하는 기법이다. 본 연구에서는 분할-정복 기법을 벌점화 분위수 회귀에 적용하고 베이즈 정보기준을 활용하여 변수를 선택하는 방법에 관하여 연구하였다. 제안 방법은 분할 수를 적절하게 선택하였을 때, 전체 데이터로 계산한 일반적인 분위수 회귀 추정량만큼 변수 선택의 측면에서 일관된 결과를 제공하면서 계산 속도의 측면에서 효율적이다. 이러한 제안된 방법의 장점은 시뮬레이션 데이터 및 실제 데이터 분석을 통해 확인하였다.

• 응용통계연구
• /
• 제34권3호
• /
• pp.411-425
• /
• 2021

가속화 실패시간모형은 로그 생존시간과 공변량간의 선형적 관계를 묘사해 준다. 가속화 실패시간모형에서 생존시간의 평균뿐만 아니라 변동성에도 영향을 미치는 공변량 효과를 추론하는 것은 흥미가 있다. 이를 위해 생존시간의 평균뿐만 아니라 분산을 모형화 하는 것이 필요하며, 이러한 모형을 평균-분산 가속화 실패시간모형이라 부른다. 본 논문에서는 벌점 가능도함수를 이용하여 평균-분산 가속화 실패시간모형에서 회귀모수에 대한 변수선택 절차를 제안한다. 여기서 벌점함수로서 LASSO, ALASSO, SCAD 그리고 HL (계층가능도)와 같은 네 가지 벌점함수를 연구한다. 제안된 변수선택 절차를 통해 중요한 공변량의 선택 뿐만 아니라 회귀모수의 추정을 동시에 제공할 수 있다. 제안된 방법의 성능은 모의실험을 통해 평가하고, 하나의 임상 예제자료를 통해 제안된 방법을 예증하고자 한다.

• Communications for Statistical Applications and Methods
• /
• 제22권2호
• /
• pp.173-180
• /
• 2015

We propose a selection procedure of principal components in principal component regression. Our method selects principal components using variable selection procedures instead of a small subset of major principal components in principal component regression. Our procedure consists of two steps to improve estimation and prediction. First, we reduce the number of principal components using the conventional principal component regression to yield the set of candidate principal components and then select principal components among the candidate set using sparse regression techniques. The performance of our proposals is demonstrated numerically and compared with the typical dimension reduction approaches (including principal component regression and partial least square regression) using synthetic and real datasets.

• Communications for Statistical Applications and Methods
• /
• 제27권2호
• /
• pp.225-239
• /
• 2020

Analysis approaches for single compositional data are well established; however, effective analysis strategies for paired compositional data remain to be investigated. The current project was motivated by studies of age-related hearing loss (presbyacusis), where subjects are classified into four audiometric phenotypes that need to be ranked within these phenotypes based on their paired compositional data. We address this challenge by formulating this problem as a classification problem and integrating a penalized multinomial logistic regression model with compositional data analysis approaches. We utilize Elastic Net for a penalty function, while considering average, absolute difference, and perturbation operators for compositional data. We applied the proposed approach to the presbyacusis study of 532 subjects with probabilities that each ear of a subject belongs to each of four presbyacusis subtypes. We further investigated the ranking of presbyacusis subjects using the proposed approach based on previous literature. The data analysis results indicate that the proposed approach is effective for ranking subjects based on paired compositional data.

• Communications for Statistical Applications and Methods
• /
• 제19권6호
• /
• pp.809-817
• /
• 2012

We propose a semiparametric inference for a generalized varying coefficient partially linear model(VCPLM) for negative binomial data. The VCPLM is useful to model real data in that varying coefficients are a special type of interaction between explanatory variables and partially linear models fit both parametric and nonparametric terms. The negative binomial distribution often arise in modelling count data which usually are overdispersed. The varying coefficient function estimators and regression parameters in generalized VCPLM are obtained by formulating a penalized likelihood through smoothing splines for negative binomial data when the shape parameter is known. The performance of the proposed method is then evaluated by simulations.

• Communications for Statistical Applications and Methods
• /
• 제26권4호
• /
• pp.359-370
• /
• 2019

Grouping structures in covariates are often ignored in regression models. Recent statistical developments considering grouping structure shows clear advantages; however, reflecting the grouping structure on the quantile regression model has been relatively rare in the literature. Treating the grouping structure is usually conducted by employing a group penalty. In this work, we explore the idea of group penalty to the quantile regression models. The grouping structure is assumed to be known, which is commonly true for some cases. For example, group of dummy variables transformed from one categorical variable can be regarded as one group of covariates. We examine the group quantile regression models via two real data analyses and simulation studies that reveal the beneficial performance of group quantile regression models to the non-group version methods if there exists grouping structures among variables.