• The Korean Journal of Applied Statistics
• /
• v.29 no.5
• /
• pp.773-786
• /
• 2016

Multiple quantile regression that simultaneously estimate several conditional quantiles of response given covariates can provide a comprehensive information about the relationship between the response and covariates. Some quantile estimates can cross if conditional quantiles are separately estimated; however, this violates the definition of the quantile. To tackle this issue, multiple quantile regression with non-crossing constraints have been developed. In this paper, we carry out a comparison study on several popular methods for non-crossing multiple linear quantile regression to provide practical guidance on its application.

• Communications for Statistical Applications and Methods
• /
• v.19 no.2
• /
• pp.293-301
• /
• 2012

Quantile regression proposed by Koenker and Bassett (1978) is a statistical technique that estimates conditional quantiles. The advantage of using quantile regression is the robustness in response to large outliers compared to ordinary least squares(OLS) regression. A regression tree approach has been applied to OLS problems to fit flexible models. Loh (2002) proposed the GUIDE algorithm that has a negligible selection bias and relatively low computational cost. Quantile regression can be regarded as an analogue of OLS, therefore it can also be applied to GUIDE regression tree method. Chaudhuri and Loh (2002) proposed a nonparametric quantile regression method that blends key features of piecewise polynomial quantile regression and tree-structured regression based on adaptive recursive partitioning. Lee and Lee (2006) investigated wage determinants in the Korean labor market using the Korean Labor and Income Panel Study(KLIPS). Following Lee and Lee, we fit three kinds of quantile regression tree models to KLIPS data with respect to the quantiles, 0.05, 0.2, 0.5, 0.8, and 0.95. Among the three models, multiple linear piecewise quantile regression model forms the shortest tree structure, while the piecewise constant quantile regression model has a deeper tree structure with more terminal nodes in general. Age, gender, marriage status, and education seem to be the determinants of the wage level throughout the quantiles; in addition, education experience appears as the important determinant of the wage level in the highly paid group.

• Journal of Preventive Medicine and Public Health
• /
• v.54 no.2
• /
• pp.145-152
• /
• 2021

Objectives: This study aimed to identify regional differences in the high-risk drinking rate among yearly alcohol users in Korea and to identify relevant regional factors for each quintile using quantile regression. Methods: Data from 227 counties surveyed by the 2017 Korean Community Health Survey (KCHS) were analyzed. The analysis dataset included secondary data extracted from the Korean Statistical Information Service and data from the KCHS. To identify regional factors related to the high-risk drinking rate among yearly alcohol users, quantile regression was conducted by dividing the data into 10%, 30%, 50%, 70%, and 90% quantiles, and multiple linear regression was also performed. Results: The current smoking rate, perceived stress rate, crude divorce rate, and financial independence rate, as well as one's social network, were related to the high-risk drinking rate among yearly alcohol users. The quantile regression revealed that the perceived stress rate was related to all quantiles except for the 90% quantile, and the financial independence rate was related to the 50% to 90% quantiles. The crude divorce rate was related to the high-risk drinking rate among yearly alcohol users in all quantiles. Conclusions: The findings of this study suggest that local health programs for high-risk drinking are needed in areas with high local stress and high crude divorce rates.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.3
• /
• pp.533-545
• /
• 2017

Quantile regression models provide a variety of useful statistical information by estimating the conditional quantile function of the response variable. However, the traditional linear quantile regression model can lead to the distorted and incorrect results when analysing real data having a nonlinear relationship between the explanatory variables and the response variables. Furthermore, as the complexity of the data increases, it is required to analyse multiple response variables simultaneously with more sophisticated interpretations. For such reasons, we propose a multivariate quantile regression tree model. In this paper, a new split variable selection algorithm is suggested for a multivariate regression tree model. This algorithm can select the split variable more accurately than the previous method without significant selection bias. We investigate the performance of our proposed method with both simulation and real data studies.

• Journal of Korea Water Resources Association
• /
• v.42 no.6
• /
• pp.465-480
• /
• 2009

In this study, single and multiple linear regression model were used to derive the relationship between precipitation and altitude, latitude and longitude in Jejudo. The single linear regression analysis was focused on whether orographic effect was existed in Jejudo by annual average precipitation, and the multiple linear regression analysis on whether orographic effect was applied to each duration and return period of quantile from regional frequency analysis by index flood method. As results of the regression analysis, it shows the relationship between altitude and precipitation strongly form a linear relationship as the length of duration and return period increase. The multiple linear regression precipitation estimates(which used altitude, latitude, and longitude information) were found to be more reasonable than estimates obtained using altitude only or altitude-latitude and altitude-longitude. Especially, as results of spatial distribution analysis by kriging method using GIS, it also provides realistic estimates for precipitation that the precipitation was occurred the southeast region as real climate of Jejudo. However, the accuracy of regression model was decrease which derived a short duration of precipitation or estimated high region precipitation even had long duration. Consequently the other factor caused orographic effect would be needed to estimate precipitation to improve accuracy.

• 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.

• Research in Community and Public Health Nursing
• /
• v.24 no.3
• /
• pp.273-281
• /
• 2013

Purpose: The purpose of this study was to determine gender differences in effects of self-efficacy, exercise benefits and barriers, and demographic factors on the physical activity. Methods: Seventy sedentary office workers, 35 male and 35 female, from a major airline company, completed a questionnaire from March 28 to April 5, 2012. Steps and body mass indices were measured using a CW-700/701 (Yamax) pedometer and Inbody 720 (Biospace), respectively. Data were analyzed using t-test, $x^2$-test, multiple linear regression, and simultaneous quantile regression. Results: For male workers, exercise self-efficacy had a significant effect on physical activity, but only when respondents were at 10%(3,431 steps/day, p=.018) and 25%(4,652 steps/day, p=.044) of the physical activity distribution. For female workers, marital status was significantly related to physical activity, but only when respondents were at 10% (3,537 steps/day, p=.013) and 25%(3,862 steps/day, p=.014) of the physical activity distribution. Conclusion: Quantile regression highlights the heterogeneous effect of physical activity determinants among office workers. Therefore intervention strategies for increasing physical activity should be tailed to genders as well as physical activity levels.

• Human Ecology Research
• /
• v.60 no.2
• /
• pp.289-302
• /
• 2022

This study examined the effect of private educational expenditure on adolescent depression and somatic symptoms. The sample comprised 2,589 first-grade middle-school students who completed the 2018 Korea Children and Youth Panel Survey. Data were analyzed using ANOVA (the generalized linear model), multiple regression, and quantile regression analysis. The principal results were as follows. First, 15.15% of adolescents reported depression symptoms, and 15.57% reported somatic symptoms. Second, levels of depression were significantly different among classes with a different level of private educational expenditure. Third, depression level was significantly negatively associated with private educational expenditure, in that the higher the private educational expenditure, the lower the depression level. Fourth, the effect of private educational expenditure on adolescent depression was significant at the 70~90th quantile regression, suggesting that private educational expenditure was associated with a higher level of depression symptoms. The results indicate that private education was viewed as a consumption commodity rather than a complementary educational practice or investment in human capital. Private education as a commodity might induce the highly developed and costly private education market. In turn, there is an increased financial burden for education at one end of the social-economic continuum and depression caused by relative deprivation at the other end.

• Journal of Korea Trade
• /
• v.26 no.1
• /
• pp.113-125
• /
• 2022

Purpose - The primary purpose of this study is to verify whether the target set out by the International Maritime Organization (IMO) for reducing carbon emissions from ships can be achieved by quantitatively analyzing the trends in technological advances of fuel oil consumption in the container shipping market. To achieve this purpose, several scenarios are designed considering various options such as eco-friendly fuels, low-speed operation, and the growth in ship size. Design/methodology - The vessel size and speed used in prior studies are utilized to estimate the fuel oil consumption of container ships and the pace of technological progress and Energy Efficiency Design Index (EEDI) regulations are added. A database of 5,260 container ships, as of 2019, is used for multiple linear regression and quantile regression analyses. Findings - The fuel oil consumption of vessels is predominantly affected by their speed, followed by their size, and the annual technological progress is estimated to be 0.57%. As the quantile increases, the influence of ship size and pace of technological progress increases, while the influence of speed and coefficient of EEDI variables decreases. Originality/value - The conservative estimation of carbon emission drawn by a quantitative analysis of the technological progress concerning the fuel efficiency of container vessels shows that it is not possible to achieve IMO targets. Therefore, innovative efforts beyond the current scope of technological progress are required.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.6
• /
• pp.1457-1469
• /
• 2017

Private education expenses is one of the key issues in Korea and there have been many discussions about it. Academically, most of previous researches for private education expenses have used multiple regression linear model based on ordinary least squares (OLS) method. However, if the data do not satisfy the basic assumptions of the OLS method such as the normality and homoscedasticity, there is a problem with the reliability of estimations of parameters. In this case, quantile regression model is preferred to OLS model since it does not depend on the assumptions of nonnormality and heteroscedasticity for the data. In the present study, the data from a survey on private education expenses, conducted by Statistics Korea in 2015 has been analyzed for investigation of the impacting factors for private education expenses. Since the data do not satisfy the OLS assumptions, quantile regression model has been employed in Bayesian approach by using gibbs sampling method. The analysis results show that the gender of the student, parent's age, and the time and cost of participating after school are not significant. Household income is positively significant in proportion to the same size for all levels (quantiles) of private education expenses. Spending on private education in Seoul is higher than other regions and the regional difference grows as private education expenditure increases. Total time for private education and student's achievement have positive effect on the lower quantiles than the higher quantiles. Education level of father is positively significant for midium-high quantiles only, but education level of mother is for all but low quantiles. Participating after school is positively significant for the lower quantiles but EBS textbook cost is positively significant for the higher quantiles.