• 대한토목학회논문집
• /
• 제26권1B호
• /
• pp.39-50
• /
• 2006

홍수를 예측하기 위해서 국내 5대강 유역의 홍수통제소는 저류함수모형을 사용하고 있으며 현재까지 홍수예측에 대한 많은 연구가 이루어지고 있다. 이에 본 논문에서는 현재 홍수통제소에서 사용되고 있는 저류함수모형과 과거의 강우-수위 관계를 이용한 회귀분석(regression analysis), 그리고 인공신경망(artificial neural network)을 이용하여 홍수를 예측하고 이를 비교, 분석하고자 하였다. 저류함수모형의 경우는 홍수통제소의 대표매개변수와 보정된 최적(평균)매개변수를 적용하였다. 그리고 회귀분석과 인공신경망은 1995~2001년까지의 홍수사상 중 4개의 홍수사상을 선택하여 회귀계수를 구하고 역전파(backpropagation) 알고리즘을 사용하여 학습을 시켰다. 그 결과 저류함수모형의 경우 최적 매개변수를 이용하였을 때 기존의 홍수통제소에서 사용하고 있는 대표매개변수보다 예측이 개선되었으며, 회귀분석의 방법인 다중회귀분석, Robust 회귀분석, Stepwise 회귀분석을 이용한 홍수예측은 비교적 정확한 결과를 얻을 수 있었다. 역전파 알고리즘을 사용한 인공신경망의 경우도 회귀분석을 이용한 홍수예측보다는 다소 못하였지만 정확한 결과를 얻을 수 있었다.

• Communications for Statistical Applications and Methods
• /
• 제27권3호
• /
• pp.365-376
• /
• 2020

Interval-valued data, a type of symbolic data, is given as an interval in which the observation object is not a single value. It can also occur frequently in the process of aggregating large databases into a form that is easy to manage. Various regression methods for interval-valued data have been proposed relatively recently. In this paper, we introduce a nonparametric regression model using the kernel function and a nonlinear regression model for the interval-valued data. We also propose applying the local linear regression model, one of the nonparametric methods, to the interval-valued data. Simulations based on several distributions of the center point and the range are conducted using each of the methods presented in this paper. Various conditions confirm that the performance of the proposed local linear estimator is better than the others.

• Journal of the Korean Data and Information Science Society
• /
• 제21권2호
• /
• pp.363-369
• /
• 2010

In this paper we propose a mixed-effects least squares support vector regression (LS-SVR) for longitudinal data. We add a random-effect term in the optimization function of LS-SVR to take random effects into LS-SVR for analyzing longitudinal data. We also present the model selection method that employs generalized cross validation function for choosing the hyper-parameters which affect the performance of the mixed-effects LS-SVR. A simulated example is provided to indicate the usefulness of mixed-effect method for analyzing longitudinal data.

• 한국지능시스템학회논문지
• /
• 제16권4호
• /
• pp.499-505
• /
• 2006

Regularlization approach to regression can be easily found in Statistics and Information Science literature. The technique of regularlization was introduced as a way of controlling the smoothness properties of regression function. In this paper, we have presented a new method to evaluate linear and non-linear fuzzy regression model based on Tanaka's model using the idea of regularlization technique. Especially this method is a very attractive approach to model non -linear fuzzy data.

• 한국농공학회지
• /
• 제22권3호
• /
• pp.75-87
• /
• 1980

Most hydro]ogic phenomena are the complex and organic products of multiple causations like climatic and hydro-geological factors. A certain significant correlation on the run-off in river basin would be expected and foreseen in advance, and the effect of each these causual and associated factors (independant variables; present-month rainfall, previous-month run-off, evapotranspiration and relative humidity etc.) upon present-month run-off(dependent variable) may be determined by multiple regression analysis. Functions between independant and dependant variables should be treated repeatedly until satisfactory and optimal combination of independant variables can be obtained. Reliability of the estimated function should be tested according to the result of statistical criterion such as analysis of variance, coefficient of determination and significance-test of regression coefficients before first estimated multiple regression model in historical sequence is determined. But some error between observed and estimated run-off is still there. The error arises because the model used is an inadequate description of the system and because the data constituting the record represent only a sample from a population of monthly discharge observation, so that estimates of model parameter will be subject to sampling errors. Since this error which is a deviation from multiple regression plane cannot be explained by first estimated multiple regression equation, it can be considered as a random error governed by law of chance in nature. This unexplained variance by multiple regression equation can be solved by stochastic approach, that is, random error can be stochastically simulated by multiplying random normal variate to standard error of estimate. Finally hybrid model on estimation of monthly run-off in nonhistorical sequence can be determined by combining the determistic component of multiple regression equation and the stochastic component of random errors. Monthly run-off in Naju station in Yong-San river basin is estimated by multiple regression model and hybrid model. And some comparisons between observed and estimated run-off and between multiple regression model and already-existing estimation methods such as Gajiyama formula, tank model and Thomas-Fiering model are done. The results are as follows. (1) The optimal function to estimate monthly run-off in historical sequence is multiple linear regression equation in overall-month unit, that is; Qn=0.788Pn+0.130Qn-1-0.273En-0.1 About 85% of total variance of monthly runoff can be explained by multiple linear regression equation and its coefficient of determination (R2) is 0.843. This means we can estimate monthly runoff in historical sequence highly significantly with short data of observation by above mentioned equation. (2) The optimal function to estimate monthly runoff in nonhistorical sequence is hybrid model combined with multiple linear regression equation in overall-month unit and stochastic component, that is; Qn=0. 788Pn+0. l30Qn-1-0. 273En-0. 10+Sy.t The rest 15% of unexplained variance of monthly runoff can be explained by addition of stochastic process and a bit more reliable results of statistical characteristics of monthly runoff in non-historical sequence are derived. This estimated monthly runoff in non-historical sequence shows up the extraordinary value (maximum, minimum value) which is not appeared in the observed runoff as a random component. (3) "Frequency best fit coefficient" (R2f) of multiple linear regression equation is 0.847 which is the same value as Gaijyama's one. This implies that multiple linear regression equation and Gajiyama formula are theoretically rather reasonable functions.

• 기술혁신학회지
• /
• 제13권2호
• /
• pp.231-251
• /
• 2010

과학기술 R&D 활동의 대표적 성과인 SCI 논문과 특허의 생산에 영향을 미치는 요인은 연구비, 연구원수, 지식스톡(R&D스톡, 논문스톡, 특허스톡 등), 연구환경, 개방화 정도, 인적자본, GDP 등 다양하다. 일반적인 회귀모형을 이용하여 논문 또는 특허의 생산에 영향을 미치는 요인을 추정하면 생산요인들 간에 다중공선성 문제가 발생하여 추정의 오류가 발생한다. 본 논문에서는 과학기술 지식생산에 영향을 미치는 요인들 간의 다중공선성 문제를 해결하기 위해 주성분 회귀모형을 이용하였다. SCI 논문을 산출로 가정한 과학생산성과와 특허를 산출로 가정한 기술생산성과에 영향을 미치는 요인을 회귀모형과 주성분 회귀모형을 이용하여 3가지 사례를 대상으로 비교 분석하였다. 일반 회귀모형을 이용하여 SCI 논문과 특허의 생산에 영향을 미치는 요인들을 분석한 결과, 요인들간에 다중공선성이 매우 높게 나타났고, 그 결과 회귀계수와 추정과 검정에 오류가 발생되었다. 반면 주성분 회귀모형을 이용하여 분석한 결과 다중공선성문제가 해결되어, 개별 생산요인에 대한 효과를 적절하게 추정할 수 있었다. 본 논문에서 제안한 주성분 회귀모형을 이용한 과학기술 지식생산함수 추정방법은 다중공선성이 강한 소수의 생산요소를 포함한 회귀분석에서 유용하게 적용될 수 있을 것이다.

• 한국수학사학회지
• /
• 제35권1호
• /
• pp.1-18
• /
• 2022

This paper presents a statistical machine learning method that generates the implied volatility surface under the rareness of the market data. We apply the practitioner's Black-Scholes model and Gaussian process regression method to construct a Bayesian inference system with observed volatilities as a prior information and estimate the posterior distribution of the unobserved volatilities. The variance instead of the volatility is the target of the estimation, and the radial basis function is applied to the mean and kernel function of the Gaussian process regression. We present two types of Gaussian process regression methods and empirically analyze them.

• 수산경영론집
• /
• 제36권1호
• /
• pp.81-101
• /
• 2005

This article is to estimate the fishing frequency function in Korean recreational fishery with respect to socio-economic characteristics of anglers. First, the study described the characteristics of the entire angler population on the view points of 9 socio-economic variables. And then, the study divided the total angler population into three groups of in-land, sea, and mixed angler populations in order to investigate the differences in their characteristics. The study could confirm the existence of differences in regions, size of regions, and educational levels between the in - land and the sea angler populations by testing heterogeneity in the frequency table. The fishing frequency function is estimated using Poisson regression model in order to accomodate the count data(non-negative discrete random variable) aspects of the fishing frequency. However, the model specification error is found due to overdispersion of data. The model exhibits the lack of goodness of fit. The negative binomial regression model is adopted to cure the overdispersion of the data as an alternative estimation methodology. Finally, the study can confirm overdispersion does not exist in the model any more and the goodness of fit improved significantly to the reasonable level. The results of estimation of fishing frequency population modeled by the negative binomial regression models are following. The three variables of region, sex, and education have effects on the decision making process of fishing frequency in the case of in-land recreation fishery. On the other hand, the three variables of sex, age, and marriage status do the same job in the case of sea angler population. Among the left-over variables, both income and use of Internet variables now affect on the process in mixed angler population. Finally, the results of whole angler population show that all of the previous variables are proven to be statistically significant due to the summation of data with all three sub-groups of angler population.

• 대한수학회보
• /
• 제57권1호
• /
• pp.51-68
• /
• 2020

In this paper, the author considers the nonparametric regression model with negatively orthant dependent random variables. The wavelet procedures are developed to estimate the regression function. For the wavelet estimator of unknown function g(·), the almost sure consistency is derived and the complete consistency is established under the mild conditions. Our results generalize and improve some known ones for independent random variables and dependent random variables.

• Journal of the Korean Data and Information Science Society
• /
• 제15권2호
• /
• pp.449-455
• /
• 2004

Support vector machine (SVM) has been very successful in pattern recognition and function estimation problems for crisp data. This paper proposes a new method to evaluate interval regression models for crisp input-output data. The proposed method is based on quadratic loss SVM, which implements quadratic programming approach giving more diverse spread coefficients than a linear programming one. The proposed algorithm here is model-free method in the sense that we do not have to assume the underlying model function. Experimental result is then presented which indicate the performance of this algorithm.