• Journal of Korea Water Resources Association
• /
• v.41 no.3
• /
• pp.325-340
• /
• 2008

This study employs Bayesian multiple regression analysis using the ordinary least squares method for regional low flow frequency analysis. The parameter estimates using the Bayesian multiple regression analysis were compared to conventional analysis using the t-distribution. In these comparisons, the mean values from the t-distribution and the Bayesian analysis at each return period are not significantly different. However, the difference between upper and lower limits is remarkably reduced using the Bayesian multiple regression. Therefore, from the point of view of uncertainty analysis, Bayesian multiple regression analysis is more attractive than the conventional method based on a t-distribution because the low flow sample size at the site of interest is typically insufficient to perform low flow frequency analysis. Also, we performed low flow prediction, including confidence interval, at two ungauged catchments in the Nakdong River basin using the developed Bayesian multiple regression model. The Bayesian prediction proves effective to infer the low flow characteristic at the ungauged catchment.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.4
• /
• pp.993-1000
• /
• 2016

Semi-supervised learning makes it easy to use an unlabeled data in the supervised learning such as classification. Applying the semi-supervised learning on the regression analysis, we propose two methods for a better regression function estimation. The proposed methods have been assumed different marginal densities of independent variables and different smoothing parameters in unlabeled and labeled data. We shows that the overfitted pilot estimator should be used to achieve the fastest convergence rate and unlabeled data may help to improve the convergence rate with well estimated smoothing parameters. We also find the conditions of smoothing parameters to achieve optimal convergence rate.

• The Korean Journal of Financial Management
• /
• v.25 no.3
• /
• pp.33-78
• /
• 2008

This paper empirically examines whether firms engage in a dynamic adjustment process toward target capital structure and, whether there is a target capital structure or mean reverting using the partial adjustment model while allowing for costly adjustment. Also we investigate the empirical determinants of optimal target capital structure in long term equilibrium. As a result, our empirical model captures at least several important features of capital structure behavior for Korean listed firms. First, Korean firms pursue target capital structure and also there is mean reverting phenomenon. Second, Non-Chaebol and small firm in adjustment speed is faster than Chaebol and large firm. Third, even capital market restricts the adjustment speed interestingly. Fourth, Korean firms have target behavior according to a degree of observed gap. Fifth, Korean firms close about one-fourth of the gap between their actual and target debt ratios within one year and thence targeting behavior explains far more of the observed changes in capital structure than market timing or pecking order considerations. Sixth, capital market is significant in determining optimal capital structure.

• The Korean Journal of Applied Statistics
• /
• v.26 no.6
• /
• pp.923-932
• /
• 2013

This paper proposes a method to predict the number of foodborne disease outbreaks by microbes. The weekly data of food poisoning occurrences by microbes in Korea contain many zero-valued observations and have dependency between outbreaks. In order to model both phenomena, the number of food poisonings is predicted by an autoregressive model and the probabilities of food poisoning occurrences by microbes (given the total of food poisonings) are estimated by the baseline category logit model. The predicted number of foodborne disease outbreaks by a microbe is obtained by multiplying the predicted number of foodborne disease outbreaks and the estimated probability of the food poisoning by the corresponding microbe. The mean squared error and the mean absolute value error are evaluated to compare the performances of the proposed method and the zero-inflated model.

• Annual Conference of KIPS
• /
• 2016.04a
• /
• pp.567-570
• /
• 2016

전력수요 예측은 설비투자, 수급 안정, 구매전력비 등에 직결되는 중요한 요소이며 국가 경제에 미치는 영향이 크다. 특히 인구가 밀집한 대도시의 경우 정치, 교육, 문화, 경제적 활동들이 전력사용과 밀접한 연관이 있어 안정적인 전력공급을 위한 정확한 전력수요 예측이 필요하다. 최근 평균기온 및 국내총생산을 독립변수로 활용하여 다중회귀모형을 구성한 연구가 전국 단위 전력수요 예측에 유용한 결과를 보여주었다. 하지만 좀 더 작은 단위 지역의 전력수요를 예측할 때에는 지역마다 제반 여건에 따른 전력사용 용도가 다르므로, 그 지역의 전력수요와 상관관계가 높은 다른 변수들을 함께 고려해야 할 필요가 있다. 본 논문은 서울시 자치구별 월 단위 전력수요 예측을 위하여 과거 전력수요량을 독립변수, 평균기온, 지역내총생산, 자치구별 인구, 세대수, 지하철 승 하차 인원을 종속변수로 설정한 다중회귀모형을 구성하였다. 이를 기반으로 다양한 실험을 통해 자치구별 월간 전력수요 예측을 진행하였으며, 그 결과 이전보다 향상된 정확도를 얻을 수 있었다.

• Journal of the Korean Data and Information Science Society
• /
• v.9 no.1
• /
• pp.19-27
• /
• 1998

We considered the bandwidth selection method which has asymptotic optimal convergence rate $n^{-1/2}$ in kernel regression function estimation. For the proposed bandwidth selection, we considered Mean Averaged Squared Error as a performance criterion and its Taylor expansion to the fourth order. Then we estimate the bandwidth which minimizes the estimated approximate value of MASE. Finally we show the relative convergence rate between optimal bandwidth and proposed bandwidth.

• The Korean Journal of Applied Statistics
• /
• v.34 no.4
• /
• pp.623-632
• /
• 2021

Through this study, we studied how to consider environment variables (such as temperatures, weekend, holiday) closely related to electricity demand, and how to consider the characteristics of Korea electricity demand. In order to conduct this study, Smoothing method, Seasonal ARIMA model and regression model with AR-GARCH errors are compared with mean absolute error criteria. The performance comparison results of the model showed that the predictive method using AR-GARCH error regression model with environment variables had the best predictive power.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.5
• /
• pp.2096-2109
• /
• 2012

This study investigates empirical issues that have received little attention in the previous research in the Korean capital market. It is to find any financial determinants on the capital structure for the firms listed in the KOSDAQ(Korea Securities Dealers Automated Quotation). Another test is performed to find any possible discriminating factors by utilizing a robust methodology, which may distinguish between the firms belonging the 'prime section' and the 'venture section' in terms of their financial aspects. Moreover, the null hypothesis that the changing trend or movement of a firm's capital structure with respect to its industry mean (or median) may be random, is also tested. For the book-value based debt ratios, size(INSIZE), growth(GROWTH), Market to book value of equity(MVBV), volatility(VOLATILITY), market value of equity (MVE) and section dummy (SECTION) showed their statistically significant effects on the book-value based leverage ratios, respectively, while size(INSIZE), growth(GROWTH), market value of equity(MVE), beta(BETA) and section dummy (SECTION) showed their statistically significant effects on the market-value based leverage ratios. This study also found an interesting result that a firm belonging to each corresponding industry has a tendency for reversion toward its mean and median leverage ratios over the five-year tested period.

• Proceedings of the Korean Statistical Society Conference
• /
• 2003.10a
• /
• pp.263-268
• /
• 2003

Breiman, Friedman, Olshen and Stone(1984)의 전체탐색법에 의한 회귀나무는 상대적으로 많은 분리가 가능한 변수로 분리기준이 정해지는 편의 현상을 갖고 있다. 본 연구에서는 이런 문제점을 해결할 수 있는 알고리즘을 제안하여 변수선택편의가 없는 회귀나무를 만들고자 한다. 제안하는 알고리즘은 노드의 분리변수를 선택하는 단계와 그 선택된 변수에 의해 이진분리를 위한 분리점을 찾는 단계로 구성되어 있다. 예측변수 중에서 목표변수와 가장 밀접하게 연관된 예측변수는 예측변수의 자료의 종류에 따라 스피어만의 순위상관계수에 의한 검정 혹은 크루스칼-왈리스의 통계량에 의한 검정을 수행하여 가장 통계적으로 유의한 변수로 선택하였고, 선택된 변수에만 Breiman et al.(1984)의 전체선택법을 적용하여 분리점을 결정하였다. 모의실험을 통해 변수선택편의, 변수선택력 , 그리고 평균제곱오차 측면에서 Breiman et al. (1984)의 CART(Classification and Regression Trees)와 제안한 알고리즘을 서로 비교하였다. 또한, 두 알고리즘을 실제 자료에 적용하여 효율을 서로 비교하였다.

• Proceedings of the Korea Society for Simulation Conference
• /
• 1998.10a
• /
• pp.170-173
• /
• 1998

본 연구에서는 시뮬레이터나 그와 유사한 가상현실환경(Virtual Reality Environment ; VRE)에서 일어날 수 있는 Simulator Sickness가 어떤 사람들에게 쉽게 발생하는지를 예측하기 위하여 다중선형회귀(Multiple linear regression) 방정식으로 예측회귀모형을 제시하였다. 이 회귀모형에서의 종속변수는 김도희 외(1998)에 의해 개발된 RSSQ의 종합점수이고, 독립변수는 실제운전경력에 1을 더한 값에 나이를 곱한 값, 과거 멀미를 경험한 정도, 1주일 평균 동화상 시간, 현재의 건강상태로 되어져 있다. 이 회귀모형의 R2값은 약 0.52로 Kolasinski(1996)의 모델보다 설명력이 18% 증가하였고, 부수적인 별도의 실험을 하지 않고도 간단한 개인 신상에 관한 간단한 자료만으로도 훨씬 좋은 결과를 예측할 수 있게 되었다. 따라서 시뮬레이터나 가상현실에서 일어나는 Simulator Sickness가 어떠한 사람에게 걸리기가 쉬운지를 쉽게 예측할 수 있게 되었고, 이러한 사람들에게는 시뮬레이터나 가상현실의 이용을 자제시키거나 주의를 주어 특별관리 함으로써 시뮬레이터나 가상현실을 운영하는데 많은 도움을 줄 수 있을 것이다.