• Communications for Statistical Applications and Methods
• /
• v.17 no.2
• /
• pp.183-191
• /
• 2010

Support vector quantile regression(SVQR) is capable of providing more complete description of the linear and nonlinear relationships among random variables. In this paper we propose an iterative reweighted least squares(IRWLS) procedure to solve the problem of SVQR with a weighted quadratic loss function. Furthermore, we introduce the generalized approximate cross validation function to select the hyperparameters which affect the performance of SVQR. Experimental results are then presented which illustrate the performance of the IRWLS procedure for SVQR.

• The Korean Journal of Applied Statistics
• /
• v.30 no.5
• /
• pp.733-745
• /
• 2017

Local composite quantile regression is a useful non-parametric regression method widely used for its high efficiency. Data smoothing methods using kernel are typically used in the estimation process with performances that rely largely on the smoothing parameter rather than the kernel. However, $L_2$-norm is generally used as criterion to estimate the performance of the regression function. In addition, many studies have been conducted on the selection of smoothing parameters that minimize mean square error (MSE) or mean integrated square error (MISE). In this paper, we explored the optimality of selecting smoothing parameters that determine the performance of non-parametric regression models using local linear composite quantile regression. As evaluation criteria for the choice of smoothing parameter, we used mean absolute error (MAE) and mean integrated absolute error (MIAE), which have not been researched extensively due to mathematical difficulties. We proved the uniqueness of the optimal smoothing parameter based on MAE and MIAE. Furthermore, we compared the optimal smoothing parameter based on the proposed criteria (MAE and MIAE) with existing criteria (MSE and MISE). In this process, the properties of the proposed method were investigated through simulation studies in various situations.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.44 no.3
• /
• pp.117-124
• /
• 2021

The development of IOT technology and artificial intelligence technology is promoting the smartization of manufacturing system. In this study, data extracted from acceleration sensor and current sensor were obtained through experiments in the cutting process of SKD11, which is widely used as a material for special mold steel, and the amount of tool wear and product surface roughness were measured. SVR (Support Vector Regression) is applied to predict the roughness of the product surface in real time using the obtained data. SVR, a machine learning technique, is widely used for linear and non-linear prediction using the concept of kernel. In particular, by applying GSVQR (Generalized Support Vector Quantile Regression), overestimation, underestimation, and neutral estimation of product surface roughness are performed and compared. Furthermore, surface roughness is predicted using the linear kernel and the RBF kernel. In terms of accuracy, the results of the RBF kernel are better than those of the linear kernel. Since it is difficult to predict the amount of tool wear in real time, the product surface roughness is predicted with acceleration and current data excluding the amount of tool wear. In terms of accuracy, the results of excluding the amount of tool wear were not significantly different from those including the amount of tool wear.

• Annals of Occupational and Environmental Medicine
• /
• v.35
• /
• pp.23.1-23.14
• /
• 2023

Background: Exposure to heavy metals is a public health concern worldwide. Previous studies on the association between heavy metal exposure and neurobehavioral functions in children have focused on single exposures and clinical manifestations. However, the present study evaluated the effects of heavy metal complex exposure on subclinical neurobehavioral function using a Korean Computerized Neurobehavior Test (KCNT). Methods: Urinary mercury, lead, cadmium analyses as well as symbol digit substitution (SDS) and choice reaction time (CRT) tests of the KCNT were conducted in children aged between 10 and 12 years. Reaction time and urinary heavy metal levels were analyzed using partial correlation, linear regression, Bayesian kernel machine regression (BKMR), the weighted quantile sum (WQS) regression and quantile G-computation analysis. Results: Participants of 203 SDS tests and 198 CRT tests were analyzed, excluding poor cooperation and inappropriate urine sample. Partial correlation analysis revealed no association between neurobehavioral function and exposure to individual heavy metals. The result of multiple linear regression shows significant positive association between urinary lead, mercury, and CRT. BMKR, WQS regression and quantile G-computation analysis showed a statistically significant positive association between complex urinary heavy metal concentrations, especially lead and mercury, and reaction time. Conclusions: Assuming complex exposures, urinary heavy metal concentrations showed a statistically significant positive association with CRT. These results suggest that heavy metal complex exposure during childhood should be evaluated and managed strictly.

• Journal of the Korean Data and Information Science Society
• /
• v.24 no.2
• /
• pp.211-222
• /
• 2013

We consider the problem of nonparametrically estimating the conditional quantile function from censored data and propose new estimators here. They are based on local logistic regression technique of Lee et al. (2006) and "double-kernel" technique of Yu and Jones (1998) respectively, which are modified versions under random censoring. We compare those with two existing estimators based on a local linear fits using the check function approach. The comparison is done by a simulation study.

• The Korean Journal of Applied Statistics
• /
• v.37 no.1
• /
• pp.13-38
• /
• 2024

As economic uncertainties have increased recently due to COVID-19, there is a growing need to quickly grasp private consumption trends that directly reflect the economic situation of private economic entities. This study proposes a method of estimating private consumption in real-time by comprehensively utilizing big data as well as existing macroeconomic indicators. In particular, it is intended to improve the accuracy of private consumption estimation by comparing and analyzing various machine learning methods that are capable of fitting ultra-high-dimensional big data. As a result of the empirical analysis, it has been demonstrated that when the number of covariates including big data is large, variables can be selected in advance and used for model fit to improve private consumption prediction performance. In addition, as the inclusion of big data greatly improves the predictive performance of private consumption after COVID-19, the benefit of big data that reflects new information in a timely manner has been shown to increase when economic uncertainty is high.