• The Korean Journal of Applied Statistics
• /
• v.10 no.2
• /
• pp.241-251
• /
• 1997

Several studies have bee down on the work of dynamic graphical methods for regression diagnostics. The main propose of the methods were to investigate (1) the effects of change of data, or (2) the effects of change of regression coefficients on the regression models. But, by contrast, we can also investigate the effects of change of regression residuals on the regression model. This method can be used in fitting better a certain set of observations to a regression model than the other observations. Our research team approaches regression diagnostics by using dynamic graphics (REDS), and we introduce REDS in this thesis.

• Journal of the Korean Data and Information Science Society
• /
• v.10 no.1
• /
• pp.173-191
• /
• 1999

In this paper we suggest the visualized regression diagnosis tool. The tool is developed by Hangul Delphi on the basis of windows, so users can easily make use of this tool though they do not have the expert knowledge about statistics and computer. Especially, to apply this tool to teaching regression analysis or data analysis, we offer various residual plots in the tool and show the results of analysis graphically.

• Korean Journal of Agricultural Science
• /
• v.39 no.3
• /
• pp.421-425
• /
• 2012

The objective of this study was to improve the detection limit of rapid detection kit for Salmonella Typhimurium by image analysis system. The rapid detection kit was comprised of four elements: sample pad, conjugate pad, nitrocellulose pad and absorbent pad. Gold nanoparticle and Salmonella antibody were used as a tag and a receptor. Salmonella antibody and goat rabbit IgG antibody were used as test and control lines on nitrocellulose membrane. The color intensity of test line began to increase from $10^5CFU/mL$ of Salmonella sample. A multiple linear regression analysis was employed to explain the relationship between predicted and measured number of Salmonella cells. The developed model could successfully predict the cell number of Salmonella with validation against extra-experimental result.

• The Korean Journal of Applied Statistics
• /
• v.15 no.1
• /
• pp.139-151
• /
• 2002

This study is concerned with model selection and diagnostics for nonlinear regression model through Bayes factor. In this paper, we use informative prior and simulate observations from the posterior distribution via Markov chain Monte Carlo. We propose the Laplace approximation method and apply the Laplace-Metropolis estimator to solve the computational difficulty of Bayes factor.

• Communications for Statistical Applications and Methods
• /
• v.18 no.2
• /
• pp.171-177
• /
• 2011

In order to construct approximation functions for real data, it is necessary to remove the outliers from the measured raw data before constructing the model. Conventionally, visualization and maximum residual error have been used for outlier detection, but they often fail to detect outliers for nonlinear functions with multidimensional input. Although the standard support vector regression based outlier detection methods for nonlinear function with multidimensional input have achieved good performance, they have practical issues in computational cost and parameter adjustments. In this paper we propose a practical approach to outlier detection using support vector regression that reduces computational time and defines outlier threshold suitably. We apply this approach to real data examples for validity.

• Communications for Statistical Applications and Methods
• /
• v.5 no.3
• /
• pp.571-580
• /
• 1998

선형회귀분석에서 새로운 변수가 모형에 추가될 때 변수변환의 필요성과 적절한 변환의 형태를 진단하는 기능이 있다고 알려져 있는 편잔차그림과 덧편잔차그림을 비선형회귀모형에 적용하고 이 그림들이 기능을 제대로 수행할 수 있는 조건을 알아보았다.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.35 no.11
• /
• pp.1111-1117
• /
• 2011

This paper proposed a fault detection and diagnosis (FDD) algorithm using nonlinear regression models, focusing especially on heat exchanger faults. This research concerned four working modes: those with no fault, evaporator fault, condenser fault, and evaporator and condenser faults. This research used no fault mode data to create an FDD algorithm. Using the no fault mode data, correlation functions for predicting the degree of superheat or subcool of heat exchangers (an evaporator and a condenser) were derived. Each correlation function has five inputs and one output. Based on these correlation functions, it is possible to predict the degree of superheat or subcool of each heat exchanger under various working conditions. The FDD algorithm was developed by comparing the predicted value and the simulation value. The FDD algorithm works well in all four working modes.

• Proceedings of the KIEE Conference
• /
• 2004.07d
• /
• pp.2534-2536
• /
• 2004

본 논문은 유전알고리즘을 이용하여 부정맥 진단의 최적화된 입력을 구성하는 방법을 제시한다. 심전도 신호의 특징을 추출하기 위해 웨이블릿 변환이 널리 사용되고 있지만, 추출된 특징들의 선택과 최적화의 문제에 대해서는 명쾌한 해결책을 제시하지 못하고 있다. 심전도 신호는 연속 웨이블릿 변환을 이용해 5레벨로 분해되었으며, 각 서브밴드에서 추출된 계수들은 부정맥 진단을 위한 특징으로 쓰이게 된다. 웨이블릿 변환을 통해 추출된 특징들(feature)은 유전자 알고리즘과 중회귀 분석을 동하여 부정맥 진단을 위한 최적화된 특징조합이 결정되었다. 본 연구를 통해 특정레벨의 어떤 계수가 부정맥 진단에 크게 영향을 미치는지 판단할 수 있었으며 입력의 차원감소는 연산시간의 축소를 가져왔고 분류정확도를 향상시켜 분류기의 성능을 증대시켰다.

• The Korean Journal of Applied Statistics
• /
• v.15 no.2
• /
• pp.233-250
• /
• 2002

Cox's proportional hazard model is highly-used for the regression analysis of survival data in various fields. Regression diagnostics for the proportional hazards model, however, is not as well-known as the diagnostics for the classical linear models and so these diagnostic methods are not used widely in our practical data analyses. For this reason, we review the residuals proposed by several authors, and investigate how to use them in assessing the model. We also provide the results and interpretation with the analysis of PBC data using S-plus 2000 program.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.11 no.6
• /
• pp.131-142
• /
• 2007

The road user cost in indirect costs as well as direct costs such as the inspection/ diagnosis cost and the repair/reinforcement cost should be considered as one of the important items in the life-cycle cost-effective design and maintenance of the bridges. To estimate the road user cost, this paper formulates the road user cost as a sum of the user delay cost and the vehicle operating cost considering the detour effect. A numerical traffic simulation and a regression analysis are performed to develop a regression model due to a time delay. The proposed regression model is applied to the generation of the maintenance strategy based on the life-cycle cost and performance, and its effectiveness and applicability is investigated. The road user cost has a great influence on establishing the maintenance strategy, and the proposed regression model could be successfully utilized to estimate the road user cost of a bridge.