• Journal of Korean Society of Forest Science
• /
• v.95 no.3
• /
• pp.370-377
• /
• 2006

This study was conducted to develop an efficient method of sampling design appropriate for the sustainable forest management. For this, data were collected in Yangpyung-Gun, Gyunggi Province based on three different sampling designs such as systematic design, systematic cluster design, and stratified cluster design. Based on evaluation statistics, the sampling designs were compared to select a sampling method fitted to sustainable forest management. It was found that the systematical cluster sampling is the most efficient sampling method in terms of feasibility for sustainable forest management. It was also recommended that the sample plots should be made as a cluster of triangle-shape. The clusters should be consisted of a main plot and three sub-plots. And the sub-plots should be arranged with a distance of 50m from the main plot in the center of cluster.

• The Journal of the Korea Contents Association
• /
• v.18 no.8
• /
• pp.525-535
• /
• 2018

If the frequency of a particular class is excessively higher than the frequency of other classes in the classification problem, data imbalance problems occur, which make machine learning distorted. Corporate bankruptcy prediction often suffers from data imbalance problems since the ratio of insolvent companies is generally very low, whereas the ratio of solvent companies is very high. To mitigate these problems, it is required to apply a proper sampling technique. Until now, oversampling techniques which adjust the class distribution of a data set by sampling minor class with replacement have popularly been used. However, they are a risk of overfitting. Under this background, this study proposes ROSE(Random Over Sampling Examples) technique which is proposed by Menardi and Torelli in 2014 for the effective corporate bankruptcy prediction. The ROSE technique creates new learning samples by synthesizing the samples for learning, so it leads to better prediction accuracy of the classifiers while avoiding the risk of overfitting. Specifically, our study proposes to combine the ROSE method with SVM(support vector machine), which is known as the best binary classifier. We applied the proposed method to a real-world bankruptcy prediction case of a Korean major bank, and compared its performance with other sampling techniques. Experimental results showed that ROSE contributed to the improvement of the prediction accuracy of SVM in bankruptcy prediction compared to other techniques, with statistical significance. These results shed a light on the fact that ROSE can be a good alternative for resolving data imbalance problems of the prediction problems in social science area other than bankruptcy prediction.

• Journal of the Korean Data and Information Science Society
• /
• v.25 no.3
• /
• pp.523-531
• /
• 2014

Design effect is often used in designing and planning sample surveys and/or in evaluating the efficiency of complex design features of the surveys. In this study, we applied Gabler et al. (2006)'s design effect model to 2013 Consumer behavior survey for food that was carried out by stratified two-stage sampling. Usability and adequacy of the design model to a real survey data are discussed and evaluated.

• The Korean Journal of Applied Statistics
• /
• v.14 no.2
• /
• pp.289-304
• /
• 2001

2001년에 실시할 $\ulcorner$국민건강.영양조사$\lrcorner$를 위한 새로운 표본설계를 하였다. 본 표본설계에서는 표본의 대표성을 높이기 위해서 기존의 표본가구수는 유지하면서 표본조사구를 증가시키고 새로운 지역 층화변수를 추가로 도입하였다. 또한 추정량의 추정오차 공식을 유도하여 추정의 신뢰성을 측정할 수 있도록 설계하였다.

• Proceedings of the Korean Statistical Society Conference
• /
• 2003.10a
• /
• pp.59-62
• /
• 2003

양곡소비량조사는 가구부문과 사업체부문에서 소비하는 주요식품의 소비량을 조사함과 동시에 주곡인 쌀의 재고량을 조사한다. 이 조사는 해마다 통계청에서 표본설계하여 실시하는 표본조사에 의존한다. 특히, 사업체 부문의 양곡소비량 조사는 응용절사법(Modified Cut-Off Sampling Method)에 의해 표본추출되어 매년 11월에 조사가 실시되고 있다. 2003년도 사업체 양곡소비량 및 재고량조사는 2001년 사업체기초통계조사와 광공업통계조사를 모집단으로 하여 22,320 대상업체 중에서 819업체를 전수조사하고 2,145 업체를 표본조사하여 전체를 추계한다.

• Proceedings of the Korean Society of Applied Pharmacology
• /
• 1995.10a
• /
• pp.73-78
• /
• 1995

표본수를 결정하는 방법에는 크게 sequential design과 fixed sample size design이 있다. Fixed sample size design은 연구를 시행하기 전에 표본수를 합리적으로 결정하고 정해진 표본내에서 연구를 진행하는 방법이며, sequential design은 연구를 진행하면서 결과의 차이가 있는가 또는 없는가에 대해 미리 정해진 한계영역을 기준으로 계속적으로 연구대상을 추출하여 연구를 진행하는 방법이다. 여기서는 많이 사용되는 fixed sample size design에 대해서만 생각하기로 한다.

• Proceedings of the KSCN Conference
• /
• 2004.05a
• /
• pp.408.2-409
• /
• 2004

조사연구에서 조사대상의 모집단을 대표할 수 있는 표본을 추출하는 것은 조사결과를 그 모집단에 일반화시키기 위해 서는 필수적이다. 국민건강영양조사에서는 확률표출방법을 이용하여 우리나라 국민의 영양실태에 관해 대표성 있는 정보를 얻고 있는 반면, 개개인의 연구자가 일부 지역 주민을 대상으로 하는 소규모의 영양조사에서는 과학적인 표출방법을 이용하여 표본을 추출하는 것이 매우 어렵기 때문에 대부분의 경우 자원자를 대상으로 하거나 임의로 표본을 선정하는 실정이다.(중략)

• Proceedings of the Korean Nuclear Society Conference
• /
• 1995.05a
• /
• pp.1029-1034
• /
• 1995

IAEA(International Atomic Energy Agency, 국제원자력기구)에서는 사찰활동 수행시, 비복원추출을 기술하는 초기 하분포(hypergeometric distribution) 대신 복원추출을 기술하는 이항분포(binomial distribution)를 사용하여 표본크기 (sample site)를 계산하여 최대 3가지 검증방법들에 할당한다. 본 연구에서는 사찰표본할당과 관련하여 PC사용이 요구되는 반복할당법인 초기하할당법, 개선된 이항할당법, 그리고 표준할당법과 포켓계산기에서 사용 가능한 근사 할당법인 개선된 이항할당근사법과 표준이항할당근사법을 비교 검토하였다.

• The Korean Journal of Applied Statistics
• /
• v.23 no.5
• /
• pp.923-932
• /
• 2010

One of the problems in designing the fifth Korean National Health and Nutrition Examination Survey(KNHNES) is the lack of an appropriate sampling frame. Due to the significant time difference, we expect eight severe sampling frame errors if we use the sampling frame obtained from the latest population and housing census that was conducted in 2005. Thus, the construction of an appropriate sampling frame for the fifth KNHNES is crucial for a successful survey. We considered the construction of a sampling frame that overcomes the limitations of the 2005 population and housing census based frame. For the construction of eight new sampling frames, we considered the use of multiple sampling frames in which the frame for the apartment households and the frame for the general households are obtained from different sources.

• The Korean Journal of Applied Statistics
• /
• v.13 no.2
• /
• pp.353-369
• /
• 2000

We systematized an stratified spatial sample design(SSSD) that uses the adequate stratification criteria such as the shapeness or the dispersion of an interesting region in a spatial population. And we proposed an adaptive searching estimation method in the SSSD to estimate the area of region of interest in two-dimensional surfaces. When wc adopt the proposed adaptive searching estimation method in SSSD, the observing sample size is more decreased than a classical sample design that all the designed sample size is observed. Nevertheless it has been shown that we can produce the moderate result but the efficiency is a slight reduced.