• The Korean Journal of Applied Statistics
• /
• v.34 no.3
• /
• pp.491-505
• /
• 2021

We consider linear regression models in high-dimensional settings (p ≫ n) and compare various classes of priors. The spike and slab prior is one of the most widely used priors for Bayesian regression models, but its model space is vast, resulting in a bad performance in finite samples. As an alternative, various continuous shrinkage priors, including the horseshoe prior and its variants, have been proposed. Although each of the above priors has been investigated separately, exhaustive comparative studies of their performance have been conducted very rarely. In this study, we compare the spike and slab prior, the horseshoe prior and its variants in various simulation settings. The performance of each method is demonstrated in terms of the regression coefficient estimation and variable selection. Finally, some remarks and suggestions are given based on comprehensive simulation studies.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.26 no.4
• /
• pp.97-115
• /
• 2023

In recent years, the structure of the population has been changing rapidly, with a declining birthrate and aging population, and the inequality of population distribution is expanding. At this point, changes in population estimation methods are required, and more accurate estimates are needed at the subregional level. This study aims to estimate the population in 2040 at the 500m grid level by applying an explainable machine learning to Busan in order to respond to this need for a change in population estimation method. Comparing the results of population estimation by applying the explainable machine learning and the cohort component method, we found that the machine learning produces lower errors and is more applicable to estimating areas with large population changes. This is because machine learning can account for a combination of variables that are likely to affect demographic change. Overestimated population values in a declining population period are likely to cause problems in urban planning, such as inefficiency of investment and overinvestment in certain sectors, resulting in a decrease in quality in other sectors. Underestimated population values can also accelerate the shrinkage of cities and reduce the quality of life, so there is a need to develop appropriate population estimation methods and alternatives.

• International Journal of Highway Engineering
• /
• v.11 no.2
• /
• pp.99-109
• /
• 2009

Fine-size exposed aggregate portland cement concrete pavements (FEACP) have surface texture of exposed aggregate by removing upper 2$\sim$3mm mortar of surface of which curing is delayed by using delay-setting agent. FEACPs have advantages of maintaining low-noise and adequate skid-resistance level during the performance period than general portland cement concrete pavements. It is necessary to ensure the durability environmental loading to prevent unexpected distress during the service life of FEACP. In the process of curing, volume change accompanied change in by moisture and temperature could be an important cause of crack in concrete to construct for successful FEACP, The use of chloride containing deicer may accelerate defects of concrete pavement, such as crack and scaling. This study aim to evaluate environmental loading resistance of FEACP, based on the estimation of shrinkage-crack-control-capability by moisture evaporation and scaling by deicer in freeze-thaw reaction.

• Journal of the Korea Concrete Institute
• /
• v.21 no.5
• /
• pp.619-628
• /
• 2009

To predict the time-dependent behavior of concrete structures, the models which describe the time-dependent characteristics of concrete, i.e. creep and shrinkage are required. However, there must be significant differences between the displacements that are obtained using the given creep and shrinkage models and the measured displacements, because of the uncertainties of creep and shrinkage model itself and those of environmental condition. There are some efforts to reduce these error or uncertainties by using the model which are obtained from creep test for the concrete in construction site. Nevertheless, the predicted values from this model may be still different from the actual values due to the same reason. This study aimed to propose a method of estimating the creep coefficient from the measured displacements of concrete structure, where creep model uncertainty factor was considered as an error factor of creep model. Numerical validation for double composite steel box and concrete beam showed desirable feasibility of the presented method. Consideration of the time-dependent characteristics of creep as one of the error factors make it possible to predict long-term behaviors of concrete structures more realistically, especially long-span PSC girder bridges and concrete cable-stayed bridges of which major problem is the geometry control under construction and maintenance.

• The Korean Journal of Applied Statistics
• /
• v.26 no.3
• /
• pp.441-452
• /
• 2013

This study analyzed the relationship between key film and a box office record success factors based on movies released in the first quarter of 2013 in Korea. An over-fitting problem can happen if there are too many explanatory variables inserted to regression model; in addition, there is a risk that the estimator is instable when there is multi-collinearity among the explanatory variables. For this reason, optimal variable selection based on high explanatory variables in box-office performance is of importance. Among the numerous ways to select variables, LASSO estimation applied by a generalized linear model has the smallest prediction error that can efficiently and quickly find variables with the highest explanatory power to box-office performance in order.

• Journal of the Institute of Convergence Signal Processing
• /
• v.11 no.4
• /
• pp.270-277
• /
• 2010

The BAMS filter is a kind of wavelet shrinkage filter based on the Bayes estimators with no simulation, therefore it can be used for a real time filter. The denoising efficiency of BAMS filter is seriously affected by the estimated noise variance in each wavelet band. To remove noise in signals in existing BAMS filter, the noise variance is estimated by using the quartile of the finest level of details in the wavelet decomposition, and with this variance, the noise of the level is removed. In this paper, to remove the image noise includingodified quartile of the level of detail is proposed. And by these techniques, the image noises of mid and high frequency bands are removed, and the results showed that the increased PSNR of ab the midband noise, the noise variance estimation method using the monotonic transform and the mout 2[dB] and the effectiveness in denosing of low noise deviation images.

• Journal of Korean Society of Steel Construction
• /
• v.24 no.5
• /
• pp.571-582
• /
• 2012

Prestressed composite girders provide efficient composite action by applying prestress to the sub-encasing concrete. In this study, an enhanced prestressed composite girder with forms suspended from the steel girder is utilized. Long-term behavior of the prestressed composite girder is estimated using age-adjusted effective modulus method, which is verified experimentally using measurements obtained from an in-service bridge. Then, parametric study is carried out to investigate the influences caused by ambient temperature, humidity, prestressing and concrete casting date.

• Proceedings of the Korean Society For Composite Materials Conference
• /
• 2005.04a
• /
• pp.245-248
• /
• 2005

During manufacturing thick composite cylinders, large thermal residual stresses are developed and induce catastrophic interlaminar failures. Since the residual stresses are dependent on many process parameters, such as temperature distribution during cure, cure shrinkage, winding tension, and migration of fibers, calculation of the residual stresses is very difficult. Therefore a radial-cut method have been used to measure the residual stresses in the composite cylinders. But the conventional radial-cut method needs to know numerous material properties which are not only troublesome to obtain but also vary with change of fiber arrangement during consolidation. In this paper, a new radial-cut method with cut-cylinder-bending test was proposed and the measured residual stresses were compared with calculated thermal residual stresses. It was found that the new radial-cut method which does not need to know any of material properties gave better estimation of residual stresses regardless of radial variation of material properties. Additionally, interlaminar tensile strength could be obtained by the cut-cylinder-bending test.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2008.05a
• /
• pp.103-107
• /
• 2008

This study was to execute the experiment for analyzing the quality characteristics of mortar by type of recycled sand by using the recycled sand produced by the manufacturing system by wet gravity separation in order to develop the removal device of impurities for the production of high-quality recycled sand. As a result, this study has sown that the mortar using the high-quality recycled sand through the manufacturing system by wet gravity separation in the fluidity property, strength property, and shrinkage property largely, compared with the mortar using low-quality recycled sand that not passed the device of sand flux. There was a tendency similar to the plane mortar. In conclusion, it was considered as various quality performances of the recycled sand were improved through the production stage of prototype.

• The Korean Journal of Applied Statistics
• /
• v.25 no.5
• /
• pp.829-835
• /
• 2012

When the input features are generated by factors in a classification problem, it is more meaningful to identify important factors, rather than individual features. The $F_{\infty}$-norm support vector machine(SVM) has been developed to perform automatic factor selection in classification. However, the $F_{\infty}$-norm SVM may suffer from estimation inefficiency and model selection inconsistency because it applies the same amount of shrinkage to each factor without assessing its relative importance. To overcome such a limitation, we propose the adaptive $F_{\infty}$-norm ($AF_{\infty}$-norm) SVM, which penalizes the empirical hinge loss by the sum of the adaptively weighted factor-wise $L_{\infty}$-norm penalty. The $AF_{\infty}$-norm SVM computes the weights by the 2-norm SVM estimator and can be formulated as a linear programming(LP) problem which is similar to the one of the $F_{\infty}$-norm SVM. The simulation studies show that the proposed $AF_{\infty}$-norm SVM improves upon the $F_{\infty}$-norm SVM in terms of classification accuracy and factor selection performance.