• Journal of the Korean Data and Information Science Society
• /
• 제24권5호
• /
• pp.1077-1088
• /
• 2013

High-dimensional data analysis arises from almost all scientific areas, evolving with development of computing skills, and has encouraged penalized estimations that play important roles in statistical learning. For the past years, various penalized estimations have been developed, and the least absolute shrinkage and selection operator (LASSO) proposed by Tibshirani (1996) has shown outstanding ability, earning the first place on the development of penalized estimation. In this paper, we first introduce a number of recent advances in high-dimensional data analysis using the LASSO. The topics include various statistical problems such as variable selection and grouped or structured variable selection under sparse high-dimensional linear regression models. Several unsupervised learning methods including inverse covariance matrix estimation are presented. In addition, we address further studies on new applications which may establish a guideline on how to use the LASSO for statistical challenges of high-dimensional data analysis.

• Communications for Statistical Applications and Methods
• /
• 제28권1호
• /
• pp.21-37
• /
• 2021

There is a direct connection between linear discriminant analysis (LDA) and linear regression since the direction vector of the LDA can be obtained by the least square estimation. The connection motivates the penalized LDA when the model is high-dimensional where the number of predictive variables is larger than the sample size. In this paper, we study the penalized LDA for a class of penalties, called the moderately clipped LASSO (MCL), which interpolates between the least absolute shrinkage and selection operator (LASSO) and minimax concave penalty. We prove that the MCL penalized LDA correctly identifies the sparsity of the Bayes direction vector with probability tending to one, which is supported by better finite sample performance than LASSO based on concrete numerical studies.

• Communications for Statistical Applications and Methods
• /
• 제27권4호
• /
• pp.445-458
• /
• 2020

The least absolute shrinkage and selection operator (LASSO) is a popular method for a high-dimensional regression model. LASSO has high prediction accuracy; however, it also selects many irrelevant variables. In this paper, we consider the moderately clipped LASSO (MCL) for the high-dimensional generalized linear model which is a hybrid method of the LASSO and minimax concave penalty (MCP). The MCL preserves advantages of the LASSO and MCP since it shows high prediction accuracy and successfully selects relevant variables. We prove that the MCL achieves the oracle property under some regularity conditions, even when the number of parameters is larger than the sample size. An efficient algorithm is also provided. Various numerical studies confirm that the MCL can be a better alternative to other competitors.

• Genomics & Informatics
• /
• 제14권4호
• /
• pp.149-159
• /
• 2016

With the success of the genome-wide association studies (GWASs), many candidate loci for complex human diseases have been reported in the GWAS catalog. Recently, many disease prediction models based on penalized regression or statistical learning methods were proposed using candidate causal variants from significant single-nucleotide polymorphisms of GWASs. However, there have been only a few systematic studies comparing existing methods. In this study, we first constructed risk prediction models, such as stepwise linear regression (SLR), least absolute shrinkage and selection operator (LASSO), and Elastic-Net (EN), using a GWAS chip and GWAS catalog. We then compared the prediction accuracy by calculating the mean square error (MSE) value on data from the Korea Association Resource (KARE) with body mass index. Our results show that SLR provides a smaller MSE value than the other methods, while the numbers of selected variables in each model were similar.

• Asian-Australasian Journal of Animal Sciences
• /
• 제27권12호
• /
• pp.1678-1683
• /
• 2014

This study considers a problem of genomic selection (GS) for adjacent genetic markers of Yorkshire pigs which are typically correlated. The GS has been widely used to efficiently estimate target variables such as molecular breeding values using markers across the entire genome. Recently, GS has been applied to animals as well as plants, especially to pigs. For efficient selection of variables with specific traits in pig breeding, it is required that any such variable selection retains some properties: i) it produces a simple model by identifying insignificant variables; ii) it improves the accuracy of the prediction of future data; and iii) it is feasible to handle high-dimensional data in which the number of variables is larger than the number of observations. In this paper, we applied several variable selection methods including least absolute shrinkage and selection operator (LASSO), fused LASSO and elastic net to data with 47K single nucleotide polymorphisms and litter size for 519 observed sows. Based on experiments, we observed that the fused LASSO outperforms other approaches.

• 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.

• Genomics & Informatics
• /
• 제14권4호
• /
• pp.187-195
• /
• 2016

In genetic association studies with high-dimensional genomic data, multiple group testing procedures are often required in order to identify disease/trait-related genes or genetic regions, where multiple genetic sites or variants are located within the same gene or genetic region. However, statistical testing procedures based on an individual test suffer from multiple testing issues such as the control of family-wise error rate and dependent tests. Moreover, detecting only a few of genes associated with a phenotype outcome among tens of thousands of genes is of main interest in genetic association studies. In this reason regularization procedures, where a phenotype outcome regresses on all genomic markers and then regression coefficients are estimated based on a penalized likelihood, have been considered as a good alternative approach to analysis of high-dimensional genomic data. But, selection performance of regularization procedures has been rarely compared with that of statistical group testing procedures. In this article, we performed extensive simulation studies where commonly used group testing procedures such as principal component analysis, Hotelling's $T^2$ test, and permutation test are compared with group lasso (least absolute selection and shrinkage operator) in terms of true positive selection. Also, we applied all methods considered in simulation studies to identify genes associated with ovarian cancer from over 20,000 genetic sites generated from Illumina Infinium HumanMethylation27K Beadchip. We found a big discrepancy of selected genes between multiple group testing procedures and group lasso.

• Computers and Concrete
• /
• 제24권1호
• /
• pp.7-17
• /
• 2019

The complex phenomenon of the bond formation in geopolymer is not well understood and therefore, difficult to model. This paper present applied statistical models for evaluating the compressive strength of geopolymer. The applied statistical models studied are divided into three different categories - linear regression [least absolute shrinkage and selection operator (LASSO) and elastic net], tree regression [decision and bagging tree] and kernel methods (support vector regression (SVR), kernel ridge regression (KRR), Gaussian process regression (GPR), relevance vector machine (RVM)]. The performance of the methods is compared in terms of error indices, computational effort, convergence and residuals. Based on the present study, kernel based methods (GPR and KRR) are recommended for evaluating compressive strength of Geopolymer concrete.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2018년도 학술발표회
• /
• pp.157-157
• /
• 2018

A reliable and accurate downscaling model which can provide climate change information, obtained from global climate models (GCMs), at finer resolution has been always of great interest to researchers. In order to achieve this model, linear methods widely have been studied in the past decades. However, nonlinear methods also can be potentially beneficial to solve downscaling problem. Therefore, this study explored the applicability of some nonlinear machine learning techniques such as neural network (NN), extreme learning machine (ELM), and ELM autoencoder (ELM-AE) as well as a linear method, least absolute shrinkage and selection operator (LASSO), to build a reliable temperature downscaling model. ELM is an efficient learning algorithm for generalized single layer feed-forward neural networks (SLFNs). Its excellent training speed and good generalization capability make ELM an efficient solution for SLFNs compared to traditional time-consuming learning methods like back propagation (BP). However, due to its shallow architecture, ELM may not capture all of nonlinear relationships between input features. To address this issue, ELM-AE was tested in the current study for temperature downscaling.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2023년도 학술발표회
• /
• pp.47-47
• /
• 2023

Drought occurs due to lack of water resources over an extended period and its intensity has been magnified globally by climate change. In recent years, drought over South Korea has also been intensed, and the prediction was inevitable for the water resource management and water industry. Therefore, drought forecasting over South Korea was performed in the current study with the following procedure. First, accumulated spring precipitation(ASP) driven by the 93 weather stations in South Korea was taken with their median. Then, correlation analysis was followed between ASP and Df4m, the differences of two pair of the global winter MSLP. The 37 Df4m variables with high correlations over 0.55 was chosen and sorted into three regions. The selected Df4m variables in the same region showed high similarity, leading the multicollinearity problem. To avoid this problem, a model that performs variable selection and model fitting at once, least absolute shrinkage and selection operator(LASSO) was applied. The LASSO model selected 5 variables which showed a good agreement of the predicted with the observed value, R²=0.72. Other models such as multiple linear regression model and ElasticNet were also performed, but did not present a performance as good as LASSO. Therefore, LASSO model can be an appropriate model to forecast spring drought over South Korea and can be used to mange water resources efficiently.