• Journal of applied mathematics & informatics
• /
• 제13권1_2호
• /
• pp.479-485
• /
• 2003

The influence of observations on the Wilks' lambda test of a linear hypothesis in multivariate regression is investigated using the local influence method. The perturbation scheme of case-weights is considered. A numerical example is given to show the effectiveness of the local influence method in identifying the influential observations.

• 한국수자원학회논문집
• /
• 제41권3호
• /
• pp.325-340
• /
• 2008

본 연구는 저수량 지역 빈도분석(regional low flow frequency analysis)을 수행하기 위하여 일반최소자승법(ordinary least squares method)을 이용한 Bayesian 다중회귀분석을 적용하였으며, 불확실성측면에서의 효과를 탐색하기 위하여 Bayesian 다중회귀분석에 의한 추정치와 t 분포를 이용하여 산정한 일반 다중회귀분석의 추정치의 신뢰구간을 비교분석하였다. 각 재현기간별 비교결과를 보면 t 분포를 이용하여 산정된 평균 추정치와 Bayesian 다중회귀분석에 의한 평균 추정치는 크게 다르지 않았다. 그러나 불확실성 측면에서 평가해볼 때 신뢰구간의 상한추정치와 하한추정치의 차이는 Bayesian 다중회귀분석을 사용한 경우가 기존 방법을 사용한 경우보다 훨씬 작은 것으로 나타났으며, 이로부터 저수량(low flow) 지역 빈도분석을 수행하는 경우 Bayesian 다중회귀분석이 일반 회귀분석보다 불확실성을 표현하는데 있어서 우수하다는 결과를 얻을 수 있었다. 또한 낙동강 유역에 2개의 미계측 유역을 선정하고 구축된 Bayesian 다중회귀모형을 적용하여 불확실성을 포함한 미계측 유역에서의 저수량(low flow)을 추정하였으며 이와 같은 방법이 미계측 유역에서의 저수(low flow) 특성을 나타내는 데 있어서 효과적일 수 있음을 입증하였다.

• 디지털융복합연구
• /
• 제18권12호
• /
• pp.481-488
• /
• 2020

기호적 회귀분석 (Symbolic Regression)은 회귀분석에서 주어진 데이터에 대하여 종속변수와 독립변수들 사이의 관계를 설명할 수 있는 함수를 직접 생성하는 분석방법으로서 Genetic Programming 이 본 분야의 연구에 가장 선도적으로 적용되고 있으며, 고정된 모델로부터 매개변수들의 최적화를 추구하는 다른 회귀분석 알고리즘들에 비하여 해석이 가능한 모델을 직접 도출할 수 있다는 장점을 갖는다. 본 연구에서는 Coarse grained 병렬 모델에 기반한 Parellel Genetic Programming 을 이용한 symbolic regression 알고리즘을 제시하고 제시된 알고리즘을 PMLB 데이타에 적용하여 해당 알고리즘의 효용성을 분석하고자 한다.

• 한국군사과학기술학회지
• /
• 제10권1호
• /
• pp.22-32
• /
• 2007

This paper describes a study of a helicopter database for the sizing stage of a preliminary design process. The database includes specifications and performance parameters for more than 150 conventional single rotor helicopters currently in market. Design parameters, including configuration and weight parameters, have been analyzed and trend curve equations(regression equations) are derived using the regression analysis method. Finally, the applicability of this research result was verified whether the method is reliable for being adopted as a useful design tool in the early stage of a helicopter design process.

• 대한산업공학회지
• /
• 제41권4호
• /
• pp.338-343
• /
• 2015

We suggest a method to configure feature functions of continuous conditional random field (C-CRF). Regression tree and similarity analysis are introduced to construct the first and second feature functions of C-CRF, respectively. Rules from the regression tree are transformed to logic functions. If a logic in the set of rules is true for a data then it returns the corresponding value of leaf node and zero, otherwise. We build an Euclidean similarity matrix to define neighborhood, which constitute the second feature function. Using two feature functions, we make a C-CRF model and an illustrate example is provided.

• Industrial Engineering and Management Systems
• /
• 제8권3호
• /
• pp.148-154
• /
• 2009

In recent days, large-sized flat-panel display (FPD) has been increasingly applied to computer monitors and TVs. Mura defects, appearing as low contrast or non-uniform brightness region, sometimes occur in manufacturing of the Thin-Film Transistor Liquid-Crystal Displays (TFT-LCD). Implementation of automatic Mura inspection methods is necessary for TFT-LCD production. Various existing Mura detection methods based on regression diagnostics, surface fitting and data transformation have been presented with good performance. This paper proposes an efficient Mura detection method that is based on a regression diagnostics using studentized residuals for automatic Mura inspection of FPD. The input image is estimated by a linear model and then the studentized residuals are calculated for filtering Mura regions. After image dilation, the proposed threshold is determined for detecting the non-uniform brightness region in TFT-LCD by means of monitoring the every pixel in the image. The experimental results obtained from several test images are used to illustrate the effectiveness and efficiency of the proposed method for Mura detection.

• Journal of Information Technology Applications and Management
• /
• 제13권4호
• /
• pp.273-290
• /
• 2006

This study suggests a frequency matrix technique to predict personal credit rate more efficiently using incomplete data sets. At first this study test on multiple discriminant analysis and logistic regression analysis for predicting personal credit rate with incomplete data sets. Missing values are predicted with mean imputation method and regression imputation method here. An artificial neural network and frequency matrix technique are also tested on their performance in predicting personal credit rating. A data set of 8,234 customers in 2004 on personal credit information of Bank A are collected for the test. The performance of frequency matrix technique is compared with that of other methods. The results from the experiments show that the performance of frequency matrix technique is superior to that of all other models such as MDA-mean, Logit-mean, MDA-regression, Logit-regression, and artificial neural networks.

• 산업경영시스템학회지
• /
• 제17권31호
• /
• pp.1-10
• /
• 1994

The purpose of this study is to search for the most effective forecasting model for condenser with independent demand among the quantitative methods such as Brown's exponential smoothing method, Box-Jenkins method, and multiple regression analysis method. The criterion for the comparison of the above models is mean squared error(MSE). The fitting results of these three methods are as follows. 1) Brown's exponential smoothing method is the simplest one, which means the method is easy to understand compared to others. But the precision is inferior to other ones. 2) Box-Jenkins method requires much historic data and takes time to get to the final model, although the precision is superior to that of Brown's exponential smoothing method. 3) Regression method explains the correlation between parts with similiar demand pattern, and the precision is the best out of three methods. Therefore, it is suggested that the multiple regression method is fairly good in precision for forecasting our item and that the method is easily applicable to practice.

• 여성건강간호학회지
• /
• 제14권3호
• /
• pp.213-221
• /
• 2008

Purpose: This study aimed to examine the accuracy and adequacy of research papers reporting statistical testings for correlation and regression. Method: Original research articles utilized correlation and regression analysis were reviewed from the Korean Journal of Women Health Nursing published from the year 2004 to 2006. Thirty-six papers were evaluated in accordance with formatted criteria in respect to an inclusiveness of research title, accuracy of statistical methods and presentation styles, and errors in reporting statistical outcomes. Result: Thirty articles (83.3%) utilized Pearson's correlational analysis, and ten articles did regression analysis. Lack of accurate understanding and interpretation of the statistical method was a main fault. Basic assumptions and diagnostic testings for each statistical method were not performed or described in most of the studies. Some points like consistency of research questions with statistical methods and criteria for sample size were still left out in part. Details of the presentation in the reporting of outcomes were not complied with the guidelines, which need careful concerns of the writers. Errors in English of result tables were found in more than one third of the tables. Conclusion: The outcome would be reflected in the submission guidelines for future writers. To reach the level comparable with internationally recognized nursing journals, concrete knowledge to apply statistical methods should be ensured in the processes of submission, reviews, and editing.

• Asian-Australasian Journal of Animal Sciences
• /
• 제21권7호
• /
• pp.923-927
• /
• 2008

The objective of this study was to compare random regression model and multiple trait animal model estimates of the (co) variance of total sperm cells over the active lifetime of AI boars. Data were provided by Smithfield Premium Genetics (Rose Hill, NC). Total number of records and animals for the random regression model were 19,629 and 1,736, respectively. Data for multiple trait animal model analyses were edited to include only records produced at 9, 12, 15, 18, 21, 24, and 27 months of age. For the multiple trait method estimates of genetic and residual variance for total sperm cells were heterogeneous among age classifications. When comparing multiple trait method to random regression, heritability estimates were similar except for total sperm cells at 24 months of age. The multiple trait method also resulted in higher estimates of heritability of total sperm cells at every age when compared to random regression results. Random regression analysis provided more detail with regard to changes of variance components with age. Random regression methods are the most appropriate to analyze semen traits as they are longitudinal data measured over the lifetime of boars.