• Health Policy and Management
• /
• v.25 no.1
• /
• pp.31-39
• /
• 2015

Background: The purpose of this study is comparison of the results between regression and multi-level analysis to find out factors influencing outcome indicators (in-hospital death, length of stay, and medical charges) of stroke patients. Methods: By using patient sample data of Health Insurance Review & Assessment Service, patients admitted with stroke were selected as survey target and 15,864 patients and 762 hospitals were surveyed. Results: For the results of existing regression analysis and multi-level analysis, models were assessed through model suitability index value and as a result, the value of results of multi-level analysis decreased compared to the results of regression, showing it is a better model. Conclusion: Factors influencing in-hospital death of stroke patients were analyzed and as a result, intra-class correlation (ICC) was 13.6%. In factors influencing length of stay, ICC was 11.4%, and medical charges, ICC was 17.7%. It was found that factors influencing the outcome indicators of stroke patients may vary in every hospital. This study could carry out more accurate analysis than existing research findings through analysis of reflecting structure at patient level and hospital level factors and analysis on random effect.

• The Korean Journal of Applied Statistics
• /
• v.31 no.1
• /
• pp.97-108
• /
• 2018

It has been routine practice in regression analysis to check the validity of the assumed model by the use of regression diagnostics tools. Outliers and influential observations often distort the regression output in an undesired manner. Jang and Anderson-Cook (Quality and Reliability Engineering International, 30, 1409-1425, 2014) proposed a graphical method (called a firework plot) so that there could be an exploratory visualization of the trace of the impact of the possible outliers and influential observations on individual regression coefficients and the overall residual sum of the squares measure. This paper further extends a graphical approach to a multi-response surface methodology problem.

• Asia-Pacific Journal of Atmospheric Sciences
• /
• v.54 no.4
• /
• pp.611-622
• /
• 2018

Observed climate data are processed under the assumption that their time series are stationary, as in multi-step temperature and precipitation prediction, which usually leads to low prediction accuracy. If a climate system model is based on a single prediction model, the prediction results contain significant uncertainty. In order to overcome this drawback, this study uses a method that integrates ensemble prediction and a stepwise regression model based on a mean-valued generation function. In addition, it utilizes empirical mode decomposition (EMD), which is a new method of handling time series. First, a non-stationary time series is decomposed into a series of intrinsic mode functions (IMFs), which are stationary and multi-scale. Then, a different prediction model is constructed for each component of the IMF using numerical ensemble prediction combined with stepwise regression analysis. Finally, the results are fit to a linear regression model, and a short-term climate prediction system is established using the Visual Studio development platform. The model is validated using temperature data from February 1957 to 2005 from 88 weather stations in Guangxi, China. The results show that compared to single-model prediction methods, the EMD and ensemble prediction model is more effective for forecasting climate change and abrupt climate shifts when using historical data for multi-step prediction.

• Korean Journal of Remote Sensing
• /
• v.38 no.5_1
• /
• pp.523-533
• /
• 2022

This paper presents a two-stage spatio-spectral fusion method (2SSFM) based on area-to-point regression kriging (ATPRK) to enhance spatial and spectral resolutions using multi-sensor satellite images with complementary spatial and spectral resolutions. 2SSFM combines ATPRK and random forest regression to predict spectral bands at high spatial resolution from multi-sensor satellite images. In the first stage, ATPRK-based spatial down scaling is performed to reduce the differences in spatial resolution between multi-sensor satellite images. In the second stage, regression modeling using random forest is then applied to quantify the relationship of spectral bands between multi-sensor satellite images. The prediction performance of 2SSFM was evaluated through a case study of the generation of red-edge and short-wave infrared bands. The red-edge and short-wave infrared bands of PlanetScope images were predicted from Sentinel-2 images using 2SSFM. From the case study, 2SSFM could generate red-edge and short-wave infrared bands with improved spatial resolution and similar spectral patterns to the actual spectral bands, which confirms the feasibility of 2SSFM for the generation of spectral bands not provided in high spatial resolution satellite images. Thus, 2SSFM can be applied to generate various spectral indices using the predicted spectral bands that are actually unavailable but effective for environmental monitoring.

• Journal of International Society for Simulation Surgery
• /
• v.3 no.1
• /
• pp.9-11
• /
• 2016

Purpose In this paper we proposed bi-directional hierarchical regression for accurate human finger landmark detection with only using depth information.Materials and Methods Our algorithm consisted of two different step, initialization and landmark estimation. To detect initial landmark, we used difference of random pixel pair as the feature descriptor. After initialization, 16 landmarks were estimated using cascaded regression methods. To improve accuracy and stability, we proposed bi-directional hierarchical structure.Results In our experiments, the ICVL database were used for evaluation. According to our experimental results, accuracy and stability increased when applying bi-directional hierarchical regression more than typical method on the test set. Especially, errors of each finger tips of hierarchical case significantly decreased more than other methods.Conclusion Our results proved that our proposed method improved accuracy and stability and also could be applied to a large range of applications such as augmented reality and simulation surgery.

• Journal of the Korea Society for Simulation
• /
• v.18 no.4
• /
• pp.137-148
• /
• 2009

The purpose of this study is to implement variable selection algorithm which helps construct a reliable linear regression model. If we use all candidate variables to construct a linear regression model, the significance of the model will be decreased and it will cause 'Curse of Dimensionality'. And if the number of data is less than the number of variables (dimension), we cannot construct the regression model. Due to these problems, we consider the variable selection problem as a combinatorial optimization problem, and apply GA (Genetic Algorithm) to the problem. Typical measures of estimating statistical significance are $R^2$, F-value of regression model, t-value of regression coefficients, and standard error of estimates. We design GA to solve multi-objective functions, because statistical significance of model is not to be estimated by a single measure. We perform experiments using simulation data, designed to consider various kinds of situations. As a result, it shows better performance than LARS (Least Angle Regression) which is an algorithm to solve variable selection problems. We modify algorithm to solve portfolio selection problem which construct portfolio by selecting stocks. We conclude that the algorithm is able to solve real problems.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2007.11a
• /
• pp.371-372
• /
• 2007

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1996.04a
• /
• pp.502-504
• /
• 1996

Plate(후판) design은 수요가 주문시 지정size(두께, 폭)로 부터 당사 압연 process를 거치면서 발생하는 지시대비 실적간의 차이를 보정하여 최종적으로 산출하게 되며, 이러한 과정은 제품생산시 size 부족으로 인한 불량 발생을 방지하는데 그 목적이 있다. Process진행중 size실적은 .gamma.-ray등 각종 측정기기로 부터 자동 측정되며 이는 process computer로 부터 main computer로 일별 전송되어 3개월 동안 조업관리 DATA BASE에 누적관리되고 있다. 본 연구는 이러한 조업실적을 근거로 제조과정에서 발생하는 size오차를 probability theory과 MULTI-REGRESSION 기법을 적용하여 DESIGN LOGIC을 개발, 제품 실수율을 향상하는데 그 목적이 있다.

• 유체기계공업학회:학술대회논문집
• /
• 2005.12a
• /
• pp.278-283
• /
• 2005

Generally Jar-Test is available to determine the coagulant dosage rate. Disadventages associated with Jar-Test are that regular samples have to be taken requiring manual intervention and the limitation to feedback control. To deal with this difficulty, determined optimized dosage rates of coagulant to investigates the union operation method of the statistical equation which uses the multi-regression method and the SCD.

• 유체기계공업학회:학술대회논문집
• /
• 2004.12a
• /
• pp.275-281
• /
• 2004

Generally Jar-Test is available to determine the coagulant dosage rate. Disadventages associated with Jar-Test are that regular samples have to be taken requiring manual intervention and the limitation to feedback control. To deal with this difficulty, determined optimized dosage rates of coagulant to Investigates the union operation method of the statistical equation which uses the multi-regression method and the SCD.