• Journal of Korean Society for Geospatial Information Science
• /
• v.20 no.3
• /
• pp.57-63
• /
• 2012

The ordinary least square (OLS) regression model is assumed that the relationship between distribution of longevity population and environmental factors to be identical. Therefore, the OLS regression analysis can't explain sufficiently the spatial characteristics of longevity phenomenon and related variables. The geographically weighted regression (GWR) model can be representing the spatial relationship of adjacent area using geographically weighted function. It also characterized which can locally explain the spatial variation of distribution of longevity population by environmental characteristics. From this point of view, this study was performed the comparative analysis between OLS and GWR model for ecological factors of longevity existing studies. In the results, GWR model has higher corresponded to model than OLS model and can be accounting for spatial variability about effect of specific environmental variables.

• Journal of the Korean Geographical Society
• /
• v.49 no.6
• /
• pp.935-948
• /
• 2014

As chronic diseases have become more prevalent and problematic, effective cares for major chronic diseases have been a locus of the healthcare policy. In this regard, this study examines how region-specific characteristics affect the prevalence of hypertension in South Korea. To analyze, we combined a unique multi-year data set including key indicators of health conditions and health behaviors at the 237 small administrative districts. The data are collected from the Annual Community Health Survey between 2009 and 2011 by Korea Centers for Disease Control and Prevention and other government organizations. For the purpose of investigating regional variations, we estimated using Geographically Weighted Regression (GWR) and decision tree model. Our finding first suggests that using the multi-year data is more legitimate than using the single-year data for the geographical analysis of chronic diseases, because the significant annual differences are observed in most variables. We also find that the prevalence of hypertension is more likely to be positively associated with the prevalence of diabetes and obesity but to be negatively associated with population density. More importantly, noticeable geographical variations in these factors are observed according to the results from the GWR. In line with this result, additional findings from the decision tree model suggest that primary influential factors that affect the hypertension prevalence are indeed heterogeneous across regional groups. Taken as a whole, accounting for geographical variations of health conditions, health behaviors and other socioeconomic factors is very important when the regionally customized healthcare policy is implemented to mitigate the hypertension prevalence. In short, our study sheds light on possible ways to manage the chronic diseases for policy makers in the local government.

• Journal of Digital Convergence
• /
• v.11 no.9
• /
• pp.255-265
• /
• 2013

The purpose of this study was to identify the variation factors of hypertension medication rate between regions and to use them as a basic data for establishment of hypertension management business plan which is customized by region. The data were collected from community health survey, National Statistics Office and National Health Insurance Corporation, and were analyzed using the geographically weighted regression. As the result of analysis, the factors that influenced the hypertension medication rate between regions were subjective recognition rate of health level, the rate of medical aid client and the number of health facility per one hundred thousand of population. According to the geographically weighted regression, the total of 230 regional regression models composed of major variables which affected the hypertension medication rate were calculated. However, this study has several limitations that the explanatory power of model is not high and others. Therefore, a follow-up study which is based on the actual data of compliance with hypertension medication will be necessary.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.14 no.3
• /
• pp.223-235
• /
• 2011

Although ordinary least squares (OLS) regression analysis can be used to examine the spatial relationships between environmental equity and urban quality of life, this global method may mask the local variations in the relationships between them. These geographical variations can not be captured without using local methods. In this context, this paper explores the spatially varying relationships between environmental equity and urban quality of life across the Atlanta metropolitan area by geographically weighted regression (GWR), a local method. Environmental equity and urban quality of life were quantified with an integrated approach of GIS and remote sensing. Results show that generally, there is a negatively significant relationship between them over the Atlanta metropolitan area. The results also suggest that the relationships between environmental equity and urban quality of life vary significantly over space and the GWR (local) model is a significant improvement on the OLS (global) model for the Atlanta metropolitan area.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.26 no.4
• /
• pp.218-236
• /
• 2023

This study attempted to analyze the environmental equity of fine dust(PM₁₀) in Daegu using MGWR(Multi-scale Geographically Weighted Regression) and KT(Korea Telecom Corporation) sensor data. Existing national monitoring network data for measuring fine dust are collected at a small number of ground-based stations that are sparsely distributed in a large area. To complement these drawbacks, KT sensor data with a large number of IoT(Internet of Things) stations densely distributed were used in this study. The MGWR model was used to deal with spatial heterogeneity and multi-scale contextual effects in the spatial relationships between fine dust concentration and socioeconomic variables. Results indicate that there existed an environmental inequity by land value and foreigner ratio in the spatial distribution of fine dust in Daegu metropolitan city. Also, the MGWR model showed better the explanatory power than Ordinary Least Square(OLS) and Geographically Weighted Regression(GWR) models in explaining the spatial relationships between the concentration of fine dust and socioeconomic variables. This study demonstrated the potential of KT sensor data as a supplement to the existing national monitoring network data for measuring fine dust.

• Journal of the Korean Data and Information Science Society
• /
• v.21 no.5
• /
• pp.831-839
• /
• 2010

Least squares support vector machine (LS-SVM) is a kernel trick gaining a lot of popularities in the regression and classification problems. We use LS-SVM to propose a iterative algorithm for a nonlinear generalized autoregressive conditional heteroscedasticity model in the mean (GARCH-M) model to estimate the mean and the conditional volatility of stock market returns. The proposed method combines a weighted LS-SVM for the mean and unweighted LS-SVM for the conditional volatility. In this paper, we show that nonlinear GARCH-M models have a higher performance than the linear GARCH model and the linear GARCH-M model via real data estimations.

• The Korean Journal of Applied Statistics
• /
• v.26 no.5
• /
• pp.697-712
• /
• 2013

When analyzing repeated binary data, the generalized estimating equations(GEE) approach produces consistent estimates for regression parameters even if an incorrect working correlation matrix is used. However, time-varying covariates experience larger changes in coefficients than time-invariant covariates across various working correlation structures for finite samples. In addition, the GEE approach may give biased estimates under missing at random(MAR). Weighted estimating equations and multiple imputation methods have been proposed to reduce biases in parameter estimates under MAR. This article studies if the two methods produce robust estimates across various working correlation structures for longitudinal binary data with time-varying covariates under different missing mechanisms. Through simulation, we observe that time-varying covariates have greater differences in parameter estimates across different working correlation structures than time-invariant covariates. The multiple imputation method produces more robust estimates under any working correlation structure and smaller biases compared to the other two methods.

• Journal of the Korean Geographical Society
• /
• v.48 no.6
• /
• pp.978-993
• /
• 2013

It can be advantage as well as disadvantage to use the spatial weight matrix in a spatial regression model; it would benefit from explicitly quantifying spatial relationships between geographical units, but necessarily involve subjective judgment while specifying the matrix. We took Incheon City as a study area and investigated how the fitness of a spatial regression model changed by constructing various spatial weight matrices. In addition, we explored neighborhood segmentation in the study area and analyzed any influence of it on the model adequacy of two basic spatial regression models, i.e., spatial lagged and spatial error models. The results showed that it can help to improve the adequacy of models to specify the spatial weight matrix strictly, that is, interpreting the neighborhood as small as possible when estimating land price. It was also found that the spatial error model would be preferred in the area with serious spatial heterogeneity. In such area, we found that its spatial heterogeneity can be alleviated by delineating sub-neighborhoods, and as a result, the spatial lagged model would be preferred over the spatial error model.

• Analytical Science and Technology
• /
• v.34 no.6
• /
• pp.241-250
• /
• 2021

Calibration curves are essential in quantitative methods and for improving the accuracy of analyte measurements in biological samples. In this study, a statistical analysis model built in the R language (The R Foundation for Statistical Computing) was used to identify a set of weighting factors and regression models based on a stepwise selection criteria. An LC-MS/MS method was used to detect the presence of urinary methamphetamine, amphetamine, and 11-nor-9-carboxy-Δ⁹ -tetrahydrocannabinol in a sample set. Weighting factors for the calibration curves were derived by calculating the heteroscedasticity of the measurements, where the presence of heteroscedasticity was determined via variance tests. The optimal regression model and weighting factor were chosen according to the sum of the absolute percentage relative error. Subsequently, the order of the regression model was calculated using a partial variance test. The proposed statistical analysis tool facilitated selection of the optimal calibration model and detection of methamphetamine, amphetamine, and 11-nor-9-carboxy-Δ⁹-tetrahydrocannabinol in urine. Thus, this study for the selection of weighting and the use of a complex regression equation may provide insights for linear and quadratic regressions in analytical and bioanalytical measurements.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.41 no.6
• /
• pp.43-51
• /
• 2013

This study aims to examine the presence of non-stationary relationship between water quality and land use in watersheds. In investigating the relationships between land use and water quality, most previous studies adopted OLS method which is assumed stationarity. However, this approach is difficult to capture the local variation of the relationships. We used 146 sampling data and land cover data of Korean Ministry of Environment to build conventional regressions and GWR models for BOD, TN and TP. Regression model and GWR models of BOD, TN, TP were compared with $R^2$, AICc and Moran's I. The results of comparisons and descriptive statistics of GWR models strongly indicated the presence of Non-Stationarity between water quality and land use.