Gyuhyeong Goh
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Gyuhyeong Goh has a total of 151 co-authors. Here are the top co-authors by number of shared publications:
- Jisang Yu
Department of Food and Resource Economics Korea University Seoul South Korea
3 shared publications Min Hua
2 shared publications- Myungjin Kim
Kyungpook National University
2 shared publications A. Salim
Department of Pathology Case Western Reserve University 2103 Cornell Road WRB Room 5105 Cleveland, OH 44106-7288
1 shared publication- Achim Goepferich
Department of Pharmaceutical Technology, University of Regensburg, 93053 Regensburg, Bavaria, Germany
1 shared publication - Ahmed El‐Ghannam
Department of Mechanical Engineering and Engineering Science, University of North Carolina at Charlotte, Charlotte, North Carolina 28223
1 shared publication
Latest Publications
- 14 October 2026
- Published by Wiley in Journal of Chemometrics
- Vol. 40 (11)
Hybrid Best Subset Selection for Bayesian Multivariate Regression
Multivariate linear regression has been widely studied and applied across many scientific fields. The rapid growth of high‐dimensional data has increased demand for fast and accurate variable selection methods in multivariate regression. This paper proposes a Bayesian best subset selection approach that efficiently searches for optimal solutions to nonconvex optimization problems arising in high‐dimensional multivariate linear regression. The proposed method employs a hybrid search strategy that integrates deterministic local optimization with stochastic global exploration, with an asymptotic global‐convergence guarantee for the fixed‐size search as the number of stochastic iterations increases. To accelerate implementation, we develop a fast computing strategy that evaluates the marginal posterior distributions of all neighboring models within a single computation. Simulation studies and real data analyses demonstrate the effectiveness and superior performance of the proposed approach.
- 14 October 2026
- Published by Wiley in Journal of Chemometrics
- Vol. 40 (11)
Hybrid Best Subset Selection for Bayesian Multivariate Regression
Multivariate linear regression has been widely studied and applied across many scientific fields. The rapid growth of high‐dimensional data has increased demand for fast and accurate variable selection methods in multivariate regression. This paper proposes a Bayesian best subset selection approach that efficiently searches for optimal solutions to nonconvex optimization problems arising in high‐dimensional multivariate linear regression. The proposed method employs a hybrid search strategy that integrates deterministic local optimization with stochastic global exploration, with an asymptotic global‐convergence guarantee for the fixed‐size search as the number of stochastic iterations increases. To accelerate implementation, we develop a fast computing strategy that evaluates the marginal posterior distributions of all neighboring models within a single computation. Simulation studies and real data analyses demonstrate the effectiveness and superior performance of the proposed approach.
- 24 April 2026
- Published by Wiley in Statistica Neerlandica
- Vol. 80 (2)
Admissibility of generalized Bayes estimators in linear regression with hierarchical g‐priors
We study the admissibility of generalized Bayes estimators for regression coefficients in normal linear models under hierarchical ‐priors. While admissibility results are available for canonical normal means problems with unknown variance, their applicability to the regression problem is nontrivial. We explore the admissibility of regression‐induced generalized Bayes estimators under scaled quadratic loss with arbitrary positive definite weight matrices by carefully transporting recent normal means results to the regression parameter. Explicit conditions on the hyperparameters governing the prior on are derived. Our results provide a decision‐theoretic justification for generalized Bayes estimators of regression coefficients under hierarchical g‐priors.
- 24 April 2026
- Published by Wiley in Statistica Neerlandica
- Vol. 80 (2)
Admissibility of generalized Bayes estimators in linear regression with hierarchical g‐priors
We study the admissibility of generalized Bayes estimators for regression coefficients in normal linear models under hierarchical ‐priors. While admissibility results are available for canonical normal means problems with unknown variance, their applicability to the regression problem is nontrivial. We explore the admissibility of regression‐induced generalized Bayes estimators under scaled quadratic loss with arbitrary positive definite weight matrices by carefully transporting recent normal means results to the regression parameter. Explicit conditions on the hyperparameters governing the prior on are derived. Our results provide a decision‐theoretic justification for generalized Bayes estimators of regression coefficients under hierarchical g‐priors.
- 6 August 2025
- Published by Elsevier in Journal of Econometrics
- Vol. 256
Bayesian space–time varying coefficient modeling for climate econometrics: A spatial–temporal Gaussian process approach
- 6 August 2025
- Published by Elsevier in Journal of Econometrics
- Vol. 256
Bayesian space–time varying coefficient modeling for climate econometrics: A spatial–temporal Gaussian process approach
- 21 April 2025
- Published by Wiley in Statistical Analysis and Data Mining: An ASA Data Science Journal
- Vol. 18 (2)
Bayesian Hybrid Model Search and Averaging for Sparse Gaussian Process Regression
Gaussian process (GP) regression has been a popular nonparametric Bayesian approach for nonlinear modeling and prediction in the fields of statistics and machine learning. However, when many predictors are considered for the construction of the kernel function, the GP approach provides unacceptable performance in both estimation and prediction. To overcome this limitation, some attempts have been made to exploit a fully Bayesian model selection approach or a penalized likelihood approach. However, the fully Bayesian framework turns out to be extremely expensive in computational terms, and the penalized likelihood method oversimplifies model uncertainties. In this paper, we propose a new sparse GP method that reduces the computational burden of fully Bayesian inference by incorporating a hybrid deterministic‐stochastic search approach into Bayesian model averaging. In addition, we develop a scalable extension of the proposed method to high‐dimensional massive data settings. The merits of the proposed methods are demonstrated via simulation experiments and real data applications.
- 21 April 2025
- Published by Wiley in Statistical Analysis and Data Mining: An ASA Data Science Journal
- Vol. 18 (2)
Bayesian Hybrid Model Search and Averaging for Sparse Gaussian Process Regression
Gaussian process (GP) regression has been a popular nonparametric Bayesian approach for nonlinear modeling and prediction in the fields of statistics and machine learning. However, when many predictors are considered for the construction of the kernel function, the GP approach provides unacceptable performance in both estimation and prediction. To overcome this limitation, some attempts have been made to exploit a fully Bayesian model selection approach or a penalized likelihood approach. However, the fully Bayesian framework turns out to be extremely expensive in computational terms, and the penalized likelihood method oversimplifies model uncertainties. In this paper, we propose a new sparse GP method that reduces the computational burden of fully Bayesian inference by incorporating a hybrid deterministic‐stochastic search approach into Bayesian model averaging. In addition, we develop a scalable extension of the proposed method to high‐dimensional massive data settings. The merits of the proposed methods are demonstrated via simulation experiments and real data applications.
- 30 March 2025
- Published by Elsevier in Statistics & Probability Letters
- Vol. 223
Posterior model consistency in high-dimensional Bayesian variable selection with arbitrary priors
- 30 March 2025
- Published by Elsevier in Statistics & Probability Letters
- Vol. 223
Posterior model consistency in high-dimensional Bayesian variable selection with arbitrary priors
A sparse empirical Bayes approach to high‐dimensional Gaussian process‐based varying coefficient models
Despite the increasing importance of high‐dimensional varying coefficient models, the study of their Bayesian versions is still in its infancy. This paper contributes to the literature by developing a sparse empirical Bayes formulation that addresses the problem of high‐dimensional model selection in the framework of Bayesian varying coefficient modelling under Gaussian process (GP) priors. To break the computational bottleneck of GP‐based varying coefficient modelling, we introduce the low‐cost computation strategy that incorporates linear algebra techniques and the Laplace approximation into the evaluation of the high‐dimensional posterior model distribution. A simulation study is conducted to demonstrate the superiority of the proposed Bayesian method compared to an existing high‐dimensional varying coefficient modelling approach. In addition, its applicability to real data analysis is illustrated using yeast cell cycle data.
A sparse empirical Bayes approach to high‐dimensional Gaussian process‐based varying coefficient models
Despite the increasing importance of high‐dimensional varying coefficient models, the study of their Bayesian versions is still in its infancy. This paper contributes to the literature by developing a sparse empirical Bayes formulation that addresses the problem of high‐dimensional model selection in the framework of Bayesian varying coefficient modelling under Gaussian process (GP) priors. To break the computational bottleneck of GP‐based varying coefficient modelling, we introduce the low‐cost computation strategy that incorporates linear algebra techniques and the Laplace approximation into the evaluation of the high‐dimensional posterior model distribution. A simulation study is conducted to demonstrate the superiority of the proposed Bayesian method compared to an existing high‐dimensional varying coefficient modelling approach. In addition, its applicability to real data analysis is illustrated using yeast cell cycle data.
- 27 May 2023
- Published by Springer Nature in Computational Statistics
- Vol. 39 (3) , 1659-1681
A hybrid deterministic–deterministic approach for high-dimensional Bayesian variable selection with a default prior
- 27 May 2023
- Published by Springer Nature in Computational Statistics
- Vol. 39 (3) , 1659-1681
A hybrid deterministic–deterministic approach for high-dimensional Bayesian variable selection with a default prior
- Open Access
- 22 July 2022
- Published by MDPI AG in Sustainability
- Vol. 14 (15) , 9020
Development of Prediction Models for Performance of Flexible Pavements in Kansas with Emphasis on the Effects of Subgrade and Unbound Layers
This study resulted from the need for better consideration of subgrade and unbound layers on the performance of flexible pavements in Kansas. Thus, the main objective was to develop pavement performance prediction models with emphasis on the effects of subgrade and unbound layers. To this end, pavement distress data, which were collected over several years across the state of Kansas, including rutting, fatigue cracking, transverse cracking, roughness and core analysis, served as the input data into statistical models. The effects of subgrade and unbound layers were represented by the corresponding results of dynamic cone penetrometer (DCP) tests and thickness of the unbound layer. In addition, traffic volume was represented by average annual daily truck traffic (AADTT). Multiple statistical analyses identified positive correlations of dynamic cone penetration index (DPI) and rate of total rutting, and DPI and percent of good core. Negative correlation was discovered between DPI and fatigue cracking code one, and DPI and percent of poor core. AADTT was positively correlated with transverse cracking codes one and two while it had no correlation with transverse cracking code zero. Thickness of the unbound layer was negatively correlated with pavement roughness and percent of poor core, while it was positively correlated with the percent of good core. Finally, the recommendation for minimum acceptable value of California bearing ratio (CBR) was provided based on the correlation between DPI and rate of change of rutting code. The recommendation enables the selection of a CBR value based on the number of years required for unit increase in the rutting code.
- 22 July 2022
- Published by MDPI AG in Sustainability
- Vol. 14 (15) , 9020
Open AccessDevelopment of Prediction Models for Performance of Flexible Pavements in Kansas with Emphasis on the Effects of Subgrade and Unbound Layers
This study resulted from the need for better consideration of subgrade and unbound layers on the performance of flexible pavements in Kansas. Thus, the main objective was to develop pavement performance prediction models with emphasis on the effects of subgrade and unbound layers. To this end, pavement distress data, which were collected over several years across the state of Kansas, including rutting, fatigue cracking, transverse cracking, roughness and core analysis, served as the input data into statistical models. The effects of subgrade and unbound layers were represented by the corresponding results of dynamic cone penetrometer (DCP) tests and thickness of the unbound layer. In addition, traffic volume was represented by average annual daily truck traffic (AADTT). Multiple statistical analyses identified positive correlations of dynamic cone penetration index (DPI) and rate of total rutting, and DPI and percent of good core. Negative correlation was discovered between DPI and fatigue cracking code one, and DPI and percent of poor core. AADTT was positively correlated with transverse cracking codes one and two while it had no correlation with transverse cracking code zero. Thickness of the unbound layer was negatively correlated with pavement roughness and percent of poor core, while it was positively correlated with the percent of good core. Finally, the recommendation for minimum acceptable value of California bearing ratio (CBR) was provided based on the correlation between DPI and rate of change of rutting code. The recommendation enables the selection of a CBR value based on the number of years required for unit increase in the rutting code.
- 18 July 2021
- Published by Taylor & Francis in Journal of Statistical Computation and Simulation
- Vol. 91 (18) , 3850-3863
Global optimal model selection for high-dimensional survival analysis
With the popularity of high-dimensional data, model selection is of great importance in recent survival analysis. In a model selection context, an important research question is how to define the best model. To answer this, various model selection criteria have been proposed for defining the best model. The existing methods commonly use the -norm penalization in order to measure the model complexity based on the number of parameters. However, due to the nonconvexity of the -penalty, finding the best model via global optimization has been a challenging research subject in statistics and machine learning. In this paper, we propose a global optimization algorithm using a modification of the simulated annealing, which is a probabilistic search algorithm for the global optimum in statistical mechanics. The performance of the proposed method is examined via simulation study and real data analysis.- 18 July 2021
- Published by Taylor & Francis in Journal of Statistical Computation and Simulation
- Vol. 91 (18) , 3850-3863
Global optimal model selection for high-dimensional survival analysis
With the popularity of high-dimensional data, model selection is of great importance in recent survival analysis. In a model selection context, an important research question is how to define the best model. To answer this, various model selection criteria have been proposed for defining the best model. The existing methods commonly use the -norm penalization in order to measure the model complexity based on the number of parameters. However, due to the nonconvexity of the -penalty, finding the best model via global optimization has been a challenging research subject in statistics and machine learning. In this paper, we propose a global optimization algorithm using a modification of the simulated annealing, which is a probabilistic search algorithm for the global optimum in statistical mechanics. The performance of the proposed method is examined via simulation study and real data analysis.- 6 May 2021
- Published by Oxford University Press (OUP) in The Econometrics Journal
- Vol. 25 (1) , 215-232
Synthetic control method with convex hull restrictions: a Bayesian maximum a posteriori approach
Synthetic control methods have gained popularity among causal studies with observational data, particularly when estimating the impacts of the interventions implemented to a small number of large units. The synthetic control methods face two major challenges: a) estimating weights for each donor to create a synthetic control and b) providing statistical inferences. To overcome these challenges, we propose a Bayesian framework that implements the synthetic control method with the parallelly shiftable convex hull and provides a Bayesian inference, which is from the duality between a penalized least squares and a Bayesian Maximum A Posteriori (MAP) approaches. Our approach differs from the recent Bayesian approaches, which allow violating the convex hull restriction and face the potential extrapolation bias. Simulation results indicate that the proposed method leads to smaller biases compared to alternatives. We revisit Abadie and Gardeazabal (2003) by applying our proposed method.
- 6 May 2021
- Published by Oxford University Press (OUP) in The Econometrics Journal
- Vol. 25 (1) , 215-232
Synthetic control method with convex hull restrictions: a Bayesian maximum a posteriori approach
Synthetic control methods have gained popularity among causal studies with observational data, particularly when estimating the impacts of the interventions implemented to a small number of large units. The synthetic control methods face two major challenges: a) estimating weights for each donor to create a synthetic control and b) providing statistical inferences. To overcome these challenges, we propose a Bayesian framework that implements the synthetic control method with the parallelly shiftable convex hull and provides a Bayesian inference, which is from the duality between a penalized least squares and a Bayesian Maximum A Posteriori (MAP) approaches. Our approach differs from the recent Bayesian approaches, which allow violating the convex hull restriction and face the potential extrapolation bias. Simulation results indicate that the proposed method leads to smaller biases compared to alternatives. We revisit Abadie and Gardeazabal (2003) by applying our proposed method.