• The Korean Journal of Applied Statistics
• /
• v.29 no.7
• /
• pp.1295-1309
• /
• 2016

This paper is to reproduce the result of Kim et al. (2014) by deriving a benchmark dose lower bound (BMDL) of lead based on the 2005 cohort data set of Children's Health and Environmental Research (CHEER) data set. The ADHD rating scales in the 2005 cohort were not consistent along the three follow-ups since two different ADHD rating scales were used in the cohort. We first unified the ADHD rating scales in the 2005 cohort by deriving a conversion formula using a penalized linear spline. We then constructed two linear mixed models for the 2005 cohort which reflected the longitudinal characteristics of the data set. The first model introduced the random intercept and the random slope terms and the second model assumed the first order autoregressive structure of the error term. Using these two models, we derived the BMDLs of lead and reconfirmed the "regression to the mean" nature of the ADHD score discovered by Kim et al. (2014). We also noticed that there was a definite difference between the sampling distributions of the two cohorts. As a result, taking this difference into account, we were able to obtain the consistent result with Kim et al. (2014).

• The Korean Journal of Applied Statistics
• /
• v.32 no.4
• /
• pp.543-559
• /
• 2019

Data often include missing values due to various reasons. If the missing data mechanism is not MCAR, analysis based on fully observed cases may an estimation cause bias and decrease the precision of the estimate since partially observed cases are excluded. Especially when data include many variables, missing values cause more serious problems. Many imputation techniques are suggested to overcome this difficulty. However, imputation methods using parametric models may not fit well with real data which do not satisfy model assumptions. In this study, we review imputation methods using nonlinear models such as kernel, resampling, and spline methods which are robust on model assumptions. In addition, we suggest utilizing imputation classes to improve imputation accuracy or adding random errors to correctly estimate the variance of the estimates in nonlinear imputation models. Performances of imputation methods using nonlinear models are compared under various simulated data settings. Simulation results indicate that the performances of imputation methods are different as data settings change. However, imputation based on the kernel regression or the penalized spline performs better in most situations. Utilizing imputation classes or adding random errors improves the performance of imputation methods using nonlinear models.

• Computers and Concrete
• /
• v.26 no.4
• /
• pp.309-316
• /
• 2020

The application of multi-variable adaptive regression spline (MARS) in predicting he long-term compressive strength of a concrete with various admixtures has been investigated in this study. The compressive strength of concrete specimens, which were made based on 24 different mix designs using various mineral and chemical admixtures in different curing ages have been obtained. First, The values of fly ash (FA), micro-silica (MS), water-reducing admixture (WRA), coarse and fine aggregates, cement, water, age of samples and compressive strength were defined as inputs to the model, and MARS analysis was used to model the compressive strength of concrete and to evaluate the most important parameters affecting the estimation of compressive strength of the concrete. Next, the proposed equation by the MARS method using particle swarm optimization (PSO) algorithm has been optimized to have more efficient equation from the economical point of view. The proposed model in this study predicted the compressive strength of the concrete with various admixtures with a correlation coefficient of R=0.958 rather than the measured compressive strengths within the laboratory. The final model reduced the production cost and provided compressive strength by reducing the WRA and increasing the FA and curing days, simultaneously. It was also found that due to the use of the liquid membrane-forming compounds (LMFC) for its lower cost than water spraying method (SWM) and also for the longer operating time of the LMFC having positive mechanical effects on the final concrete, the final product had lower cost and better mechanical properties.

• Asian Journal for Public Opinion Research
• /
• v.10 no.3
• /
• pp.200-229
• /
• 2022

The purpose of this study is to analyze the effects of youth job policy measures, set forth in Korea's 2016-2020 Third Basic Plan for Low Fertility and Aging Society (December 2015), on marriage and fertility among young people. Based on the results, we provide theoretical explanations for the findings and suggest policy alternatives to overcome the low fertility phenomenon in Korea. Previous studies have shown that employment is an important factor for marriage among youth, and a job policy could increase marriage and fertility rates. To test this assumption, we performed an exact matching between Statistics Korea's Employee-Enterprise Linkage DB and the Newlyweds DB from 2011 to 2019, in order to identify all young people aged 15-34. Then, linear spline regression analysis was used to examine the impact of the youth job policy on marriage and fertility. Comparing the period before the implementation of the employment policy (2011-2015) and after (2016-2019), the fertility rate increased as the number of young people looking for work increased. In addition, it was found that these impacts were greater after the implementation of the measures (2016-2019) than before (2011-2015). It is interesting to note that job growth among young people did not lead to an increase in marriage. However, the number of births significantly increased when young people who occupy jobs got married, which seems to be related to the delay in marriage among young people who are employed. Survey results about the intentions to marry and views on fertility are utilized for the explanation of the study results.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.414-414
• /
• 2017

토양수분은 지면환경에서 일어나는 수문 및 에너지 순환을 이해하는 데 있어 중요한 기상인자이다. 토양수분 현장관측은 땅속에 매설된 센서에 의해 상당히 정확하게 이루어지만, 관측점 수가 충분치 않아 공간적 연속성을 확보하지 못하는 어려움이 존재한다. 이에 광역적 및 연속적 관측이 가능한 마이크로파 위성센서가 토양수분 정보 획득을 위한 보조수단으로서 그 중요성이 부각되고 있다. 마이크로파 위성센서는 구름 등 기상조건의 제약을 받지 않으며, 1978년 이래 현재까지 여러 위성에 의해 25 km 및 10 km 해상도의 전지구 토양수분자료가 생산되어 왔다. 마이크로파 센서를 이용한 토양수분자료는 동일지점에 대하여 하루 2회 정도 산출되므로 적절한 시간분해능을 가지지만, 공간해상도가 최고 10 km로서 지역규모의 수문분석에 적용하기에는 충분치 않다. 이러한 토양수분자료의 공간해상도 문제 해결을 위하여 다양한 지면환경요소를 활용한 통계적 다운스케일링이 대안으로 제시되었다. 최근의 선행연구들은 대부분 방정식을 이용한 결합모형을 통해 통계적 다운스케일링을 수행하였는데, 회귀식과 같은 선형결합뿐 아니라 신경망이나 기계학습 등의 비선형결합에서도, 불가피하게 발생할 수밖에 없는 잔차(residual)로 인하여 다운스케일링 전후의 공간분포 패턴이 달라져버리는 문제를 안고 있었다. 회귀분석에 잔차의 공간내삽을 결합시킨 회귀크리깅(regression kriging)은 잔차보정을 통해 이러한 문제를 해결함으로써 다운스케일링 전후의 공간분포 일관성을 보장하는 기법이다. 이 연구에서는 회귀크리깅을 이용하여 일자별 AMSR2(Advanced Microwave Scanning Radiometer 2) 토양수분 자료를 10 km에서 1 km 해상도로 다운스케일링하고, 다운스케일링 전후의 자료패턴 일관성을 평가한다. 지면온도(LST), 지면온도상승률(RR), 식생온도건조지수(TVDI)는 일자별로 DB를 구축하였고, 식생지수(NDVI), 수분지수(NDWI), 지면알베도(SA)는 8일 간격으로 DB를 구축하였다. 이러한 8일 간격의 자료를 일자별로 변환하기 위하여 큐빅스플라인(cubic spline)을 이용하여 시계열내삽을 수행하였다. 또한 상이한 공간해상도의 자료는 최근린법을 이용하여 다운스케일링 목표해상도인 1 km에 맞도록 변환하였다. 우선 저해상도 스케일에서 추정치를 산출하기 위해서는 저해상도 픽셀별로 이에 해당하는 복수의 고해상도 픽셀을 평균화하여 대응시켜야 하며, 이를 통해 6개의 설명변수(LST, RR, TVDI, NDVI, NDWI, SA)와 AMSR2 토양수분을 반응변수로 하는 다중회귀식을 도출하였다. 이식을 고해상도 스케일의 설명변수들에 적용하면 고해상도 토양수분 추정치가 산출되는데, 이때 추정치와 원자료의 차이에 해당하는 잔차에 대한 보정이 필요하다. 저해상도 스케일로 존재하는 잔차를 크리깅 공간내삽을 통해 고해상도로 변환한 후 이를 고해상도 추정치에 부가해주는 방식으로 잔차보정이 이루어짐으로써, 다운스케일링 전후의 자료패턴 일관성이 유지되는(r>0.95) 공간상세화된 토양수분 자료를 생산할 수 있다.

• Journal of Preventive Medicine and Public Health
• /
• v.56 no.3
• /
• pp.282-290
• /
• 2023

Objectives: This study aimed to evaluate the potential interaction between kidney function and the non-linear association between serum calcium levels and cardiovascular disease (CVD) mortality. Methods: This study included 8927 participants enrolled in the Dong-gu Study. Albumin-corrected calcium levels were used and categorized into 6 percentile categories: <2.5th, 2.5-25.0th, 25.0-50.0th, 50.0-75.0th, 75.0-97.5th, and >97.5th. Restricted cubic spline analysis was used to examine the non-linear association between calcium levels and CVD mortality. Cox proportional hazard regression was used to estimate hazard ratios (HRs) for CVD mortality according to serum calcium categories. All survival analyses were stratified by the estimated glomerular filtration rate. Results: Over a follow-up period of 11.9±2.8 years, 1757 participants died, of whom 219 died from CVD. A U-shaped association between serum calcium and CVD mortality was found, and the association was more evident in the low kidney function group. Compared to the 25.0-50.0th percentile group for serum calcium levels, both low and high serum calcium tended to be associated with CVD mortality (<2.5th: HR, 6.23; 95% confidence interval [CI], 1.16 to 33.56; >97.5th: HR, 2.56; 95% CI, 0.76 to 8.66) in the low kidney function group. In the normal kidney function group, a similar association was found between serum calcium levels and CVD mortality (<2.5th: HR, 1.37; 95% CI, 0.58 to 3.27; >97.5th: HR, 1.65; 95% CI, 0.70 to 3.93). Conclusions: We found a non-linear association between serum calcium levels and CVD mortality, suggesting that calcium dyshomeostasis may contribute to CVD mortality, and kidney function may modify the association.

• The Korean Journal of Ecology
• /
• v.23 no.1
• /
• pp.25-32
• /
• 2000

To analyze the relationship between climatic factors (monthly temperatures and precipitations) and the radial growths or Pinus densiflora with different topographical settings in Worak National Park, Korea, 20 stands were chosen and 10 trees were selected from each stand. After crossdating, each ring-width series was double detrended (standardized) by fitting first a negative exponential or straight regression line and secondly a 60-year cubic spline. The growth patterns coud be categorized by four groups using cluster analysis. Cluster Ⅰ stand has north aspect, but others have south or southwest aspects. Cluster Ⅰ (one), cluster Ⅱ (ten), and cluster Ⅲ (two) stands are located in lower. elevation (305∼580 m), however, cluster Ⅳ (seven) stands are located in higher elevation, mostly in 560～870 m. Cluster Ⅱ and Ⅲ stands are located at similar elevation with the same aspect, however, cluster Ⅱ stands are located on more rocky and stiff slope with shallow soil depth. The response functions were used to examine the difference in the relationships between climatic factors and tree growths among the 4 cluster chronologies. The climatic factors are not limiting the growth in the cluster Ⅰ stand as highly as in other cluster plots because of rather mesic conditions in the north slope. The precipitation in the spring appears to be the main limiting factor in the cluster Ⅱ stands. The topographical characteristics of the sites of cluster Ⅱ, shallow soil depths on the rocky slope in the south aspect at lower elevation, may enhance the sensitivity of growth to moisture stress. In cluster Ⅲ and cluster Ⅳ, winter and spring temperature prior to the growth become more important than for cluster Ⅱ. This pattern is com-mon for Pinus densiflora trees growing in higher. elevation (equation omitted 800 m) in South Korea. It nay be re-lated with preconditioning effects of temperature as the temperature decreases with increasing elevation (cluster Ⅳ) or in the valley (cluster Ⅲ). The results obtained by tree-ring analysis were digitalized by GIS and spatio-temporal information on tree-ring data and topographic setting were analyzed and displayed simultaneously. The results of this study can be used to predict the future change of Pinus densiflora ecosystem to climate change expected in central Korea.