• Proceedings of the Korea Water Resources Association Conference
• /
• 2016.05a
• /
• pp.286-286
• /
• 2016

기후변화를 고려한 위한 미래 수자원 계획은 신뢰성 있는 수문기상인자의 산정을 통한 수자원 영향 평가 결과로 수립되는 것이 중요하다. 본 연구에서는 DHSVM모형과 TOPLATS모형에서 생산된 결과를 가지고 제약조건을 가지는 다중선형회귀 모형을 통하여 2012년-2014년 동안의 한반도 유역에 대한 수문기상인자를 산정하였다(Fig. 1). 다중선형회귀 모형은 하나의 종속변수의 변화를 설명하기 위하여 두 개 이상의 독립변수를 사용하는 모형으로 일반적으로 다중선형회귀 모형의 회귀 계수는 음의 값을 가질 수 있으므로 본 연구의 적용을 위하여 검정지점에 대하여 산정된 음의 회귀계수 값이 그대로 적용될 경우 적합하지 않으므로 회귀 계수에 제약조건을 부여하였다. 제한된 회귀 계수의 범위는 0-1사이를 가진다. 동적 다중선형 모형의 구성은 광릉 GCK, GDK 지점자료를 활용하였다.

• The Korean Journal of Applied Statistics
• /
• v.12 no.1
• /
• pp.109-124
• /
• 1999

일반적으로 오차항이 자기상관되어 있는 선형회귀 모형에서는 회귀계수에 대한 보통최소제곱추정량이 효율적이지 못 하다고 알려져 있다. 그러나 이러한 일반화선형회귀모형에서 독립변수의 형태에 따라서는 OLSE의 사용 가능성을 제시하는 모형이 있다. 본 연구에서는 오차항이 일차 이동평균 과정을 따르는 선형회귀모형에서 여러 추정량들 (GLSE, APX, MAPX)에 대한 OLSE의 상대효율함수를 유도하고 비교 분석하고자 한다. 특히 소표본에서 정확한 상대효율값을 구하여 OLSE의 효율성이 크게 떨어지지 않거나 효율성이 나은 회귀모형들을 제시한다.

• Communications for Statistical Applications and Methods
• /
• v.4 no.2
• /
• pp.333-343
• /
• 1997

선형회귀분석에서 변수의 선택문제는 최적의 모형을 찾는데 아주 중요한 부분을 차지한다. George와 McCulloch(1993)는 계층적 베이즈 모형과 깁스표본법을 이용하여 선형회귀모형에서 변수를 선택하는 문제를 고려하였다. 이 논문에서는 George와 McCulloch의 모형을 바탕으로 각각의 설명변수가 모형에 포함될 사전확률을 객관적인 기준에 의하여 결정하는 문제를 고려하여 보았다.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.38 no.6
• /
• pp.801-808
• /
• 2018

The learning of the linear regression model is stable on the assumption that the sample size is sufficiently larger than the number of explanatory variables and there is no serious multicollinearity between explanatory variables. In this study, we investigated the difficulty of model learning when the assumption was violated by analyzing a real heavy rain damage data and we proposed to use a principal component regression model or a ridge regression model after integrating data to overcome the difficulty. We evaluated the predictive performance of the proposed models by using the test data independent from the training data, and confirmed that the proposed methods showed better predictive performances than the linear regression model.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.3
• /
• pp.547-557
• /
• 2017

The health-related quality of life data are commonly skewed and bounded with spike at the perfect health status, and the variance tended to be heteroscedastic. In this study, we have developed a prediction model for EQ-5D using linear regression model, beta regression model, and extended beta regression model with mean and precision submodel, and also compared the predictive accuracy. The extended beta regression model allows to model skewness and differences in dispersion related to covariates. Although the extended beta regression model has higher prediction accuracy than the linear regression model, the overlapped confidence intervals suggested that the extended beta regression model was superior to the linear regression model. However, the expended beta regression model could explain the heteroscedasticity and predict within the bounded range. Therefore, the expended beta regression model are appropriate for fitting the health-related quality of life data such as EQ-5D.

• The Korean Journal of Applied Statistics
• /
• v.4 no.2
• /
• pp.117-127
• /
• 1991

Coefficient of determination $R^2$ is most frequently used descriptive measure in practical use of linear regression analysis. But there have been controversies on defining this measure in the cases of linear regression without the intercept, weighted linear regression and robust linear regression. Several authors such as Kvalseth(1985) and Willet and Singer(1988) proposed many variations of $R^2$ to meet the situations. However, theire measures are not satisfactory due to the lack of a universal principle. In this study, we propose a unfied approach to defining the coefficient of determination $R^2$ using the concept of likelihood distance. This new measure is in good accordance with typical $R^2$ in linear regression and, moreover, can be applied to nonlinear regression models and generalized linear models such as logit and log-linear models.

• Journal of the Korean Data and Information Science Society
• /
• v.21 no.1
• /
• pp.21-32
• /
• 2010

Graphical regression is a paradigm for obtaining regression information using plots without model assumptions. The general goal of this approach is to find lowdimensional sufficient summary plots without loss of important information. Model assessments using residual plots are less likely to be successful in models that are not linear. As an alternative approach, marginal model plots provide a general graphical method for assessing the model. We apply the methods of graphical regression and model assessment using marginal model plots to the logistic regression model.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.3
• /
• pp.475-483
• /
• 2009

In this paper we introduced a class of nonlinear time series models to analyse KOSPI data. We introduce the Generalized Power-Transformation TGARCH (GPT-TGARCH) model and the model includes Zakoian (1993) and Li and Li (1996) models as the special cases. We showed the effectiveness and efficiency of the new model based on KOSPI data.

• 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.

• 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.