• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.161-161
• /
• 2017

본 연구는 효율적인 매개변수 추정 방법인 Isolated-Speciation Particle Swarm Optimization(ISPSO)와 불확실도 분석 기법인 GLUE를 결합한 ISPSO-GLUE의 개념을 도입하였다. 임의 매개변수 추출을 방식인 GLUE 기법과 ISPSO-GLUE와의 효율성 비교를 위해 분포형 강우-유출모형인 TOPMODEL에 적용하였으며, 추정된 매개변수에 대한 모의 유량치를 이용하여 성능을 비교하였다. 연구대상지는 Texas의 $1000{\times}2000km^2$ 크기 내외의 두 유역을 택하였으며, 2002-2007년을 보정기간으로 하고, 2008-2013년을 검증기간으로 설정하였다. 불확실도 분석에 10개의 TOPMODEL 매개변수를 이용하고, 우도함수로는 Nash-Sutcliffe(NS) Coefficient이용하여 0.6이상 기준으로 행동모형을 구분하였다. 분석 결과 모수 추정성능면에서, 누적 최대 NS 값과 행동 모형의 개수는 전반적으로 ISPSO-GLUE에서 큰 값을 보였으나, 불확실도 구간에 속하는 관측치는 GLUE에서 더 높은 경향을 보였다. 이는 ISPSO-GLUE의 편향된 모수 추정으로 불확실도 구간이 작아지며, 포함되는 관측치가 GLUE에 비하여 적게 되는 것으로 확인되었다. ISPSO-GLUE의 개선을 통하여 TOPMODEL에 대한 적용성을 증진시키고, 값비싼 수문모형에 대한 매개변수 추정에 더 큰 효과를 가져올 것으로 기대된다.

• The Korean Journal of Financial Management
• /
• v.19 no.2
• /
• pp.211-231
• /
• 2002

본 연구의 목적은 VECM(Vector Error Correction Model)과 인공지능모형(Artificial Neural Networks)을 이용하여 우리나라 증권시장과 거시경제 변수들과의 장기적 관계에 대한 설명력을 비교해보고자 함에 있다. VECM이 APT(Arbitrage Pricing Theory)에 기초를 둔 선형동학모형이라고 한다면, 인공지능모형은 비모수적 비선형모형이라는 점에서, 두 방법론의 분석결과를 직접 비판하는 것은 의미있는 연구라고 할 수 있다. 인공지능모형을 주로 활용하는 선행연구들에 의하면, 증권시장은 시장의 특이패턴들로 인해 계량경제학적 접근인 선형 모형보다는 인공지능모형을 통해 증권시장의 움직임을 설명하고 예측하는 것이 더 바람직할 수도 있다는 것이다. 따라서, 본 연구에서는 VECM분석에서 자료의 안정성을 검증하고, 공적분 백터를 발견한 이후, 장기적 균형관계의 실증적 분석을 하였다. 그리고, 인공지능모형에서는 delta rule과 Sigmoid 함수를 이용한 GRNN(General Regression Neural Net)과 Back-Propagation등의 방법들을 활용하였다. 이러한 분석결과, Back-Propagation 모형이 다른 모든 모형들보다도 더 우수한 설명력을 보여주고 있었다. 이러한 결과들은 인공지능모형이 동태적인 선형 모형보다도 더 우수한 설명력을 제공할 수 있는 가능성을 보여주고 있었다.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 2002.02a
• /
• pp.375-380
• /
• 2002

장 등(2001)에 의해 개발된 보급형 홍삼제조용 건조기를 이용하여 최고급 품질의 천지삼(H&EG)등급 홍삼의 수율을 15%이상으로 향상시킬 수 있는 공정 개발을 위한 실험을 수행한 결과, 1차 건조시 최적의 상대습도 조건은 60-70%인 것으로 나타났고, 초음파 속도 330-360m/s를 가지는 원료수삼이 홍삼으로 제조된 후 천지삼 수율이 높았으며 내공과 내백도 적게 나타났다. 한편 저장기간별로 우수 홍삼의 제조결과를 보면 최고급 품질의 홍삼제조를 위해서는 수확 후 3일 이내에 제조작업을 수행하는 것이 적절할 것으로 판단되었다. 한편 본 연구에서 개발된 최고급품질의 홍삼제조공정의 천지삼 수율측정을 위해 실험을 수행한 결과 천지삼의 평균수율이 18.8%나타났다. 이 결과를 통계분석한 결과에 의하면, 95%의 신뢰를 가지는 수율구간은 14.1%-23.7%로 나타났다. 또한 상기 결과를 비모수 X$^2$검증한 결과 모집단을 대표할 수 있는 것으로 나타났다. 그러므로 본 연구과제에 의해 개발된 최고급 품질의 홍삼제조공정기술은 성공적으로 개발된 것으로 판단되었다.

• Journal for History of Mathematics
• /
• v.34 no.1
• /
• pp.21-37
• /
• 2021

The purpose of the present study is to compare the differences in levels per group of high school students regarding the self-regulated learning factors for mathematics. For this purpose, a self-regulated learning measurement tool was developed and surveys were conducted. And the statistical analysis was completed using the frequency analysis, Kolmogorov-Smirnov normality test, Mann-Whitney U test and the Kruskal-Wallis H test. As a result, it is found that self-efficacy is of statistically significant differences in self-regulated learning levels regardless of the group classifications but test anxiety does not show statistically significant differences in self-regulated learning levels regardless of the group classifications.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.16 no.6
• /
• pp.157-175
• /
• 2021

This study categorized 3,214 companies out of the tech firms supported by the Korea Technology Finance Corporation's "technology guarantee scheme" through technology assessment from 2017 to 2019 into Fourth Industrial Revolution-related companies and general SMEs. The impact of the management characteristics of these 1,752 tech firms on the determination of high-growth firms was then empirically analyzed. This study used the OECD(2007) definition to define a "high-growth firm" as "an enterprise with average revenue growth greater than 20% per annum, over a two-year period." As the two sample groups showed non-normal distribution, this study conducted the Mann-Whitney U test, a nonparametric test, to analyze the mean differences and bivariate logistic regression in which the normality assumption is less stringent. The independent variables include fundamental characteristics; a regional dummy; a technological level dummy; and the capabilities of company representatives, human capital, and technological innovation. The corresponding sub-variables are representatives' level of education and experience in the same industry, full-time workers, research personnel, the extent of intellectual property rights, investment in research and development, firm age, total assets, region_metropolitan area, region_central region, technological level_high technology, and technological level_medium technology. As a result, the research hypothesis about representatives' level of experience in the same industry, full-time workers, total assets, and technological level_high technology was supported for the Fourth Industrial Revolution-related companies. For the general SMEs, the research hypothesis about representatives' level of experience in the same industry, research personnel, total assets, and region_metropolitan area was supported.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.32 no.1
• /
• pp.55-63
• /
• 2019

In engineering problems, many random variables have correlation, and the correlation of input random variables has a great influence on reliability analysis results of the mechanical systems. However, correlated variables are often treated as independent variables or modeled by specific parametric joint distributions due to difficulty in modeling joint distributions. Especially, when there are insufficient correlated data, it becomes more difficult to correctly model the joint distribution. In this study, multivariate kernel density estimation with bounded data is proposed to estimate various types of joint distributions with highly nonlinearity. Since it combines given data with bounded data, which are generated from confidence intervals of uniform distribution parameters for given data, it is less sensitive to data quality and number of data. Thus, it yields conservative statistical modeling and reliability analysis results, and its performance is verified through statistical simulation and engineering examples.

• The Journal of the Korea Contents Association
• /
• v.12 no.7
• /
• pp.497-509
• /
• 2012

The purposed of this study was to evaluate the effects of psycho-social maturity personality improvement program for university students. This program based on the book written by Lee et al.(2010) to improve the personality for university student was used. The program consisted of a total of 13 session and every session consisted of a 50 minutes lecture and 100minutes activity program. The researcher carried out the pre-test, post test and follow up test regarding psycho-social maturity and conducted Mann-Whitney U test and the ancova. The results showed who had participated in counseling groups reported significant improvement of psycho-social maturity total scores and sub scores, while the control group did not report any significant improvements. After 10 weeks follow-up point, these changes experienced by the experiment group on the improvement of the psycho-social maturity. Finally the personality program helped improve the psycho-social maturity.

• Korean Journal of Construction Engineering and Management
• /
• v.15 no.5
• /
• pp.103-114
• /
• 2014

Data envelopment analysis (DEA) measures the relative efficiency of decision making units (DMUs) with multiple performance factors that are grouped into outputs and inputs. DEA has proven to be superior to simple aggregation of performance measures, and is also useful for evaluating the performance of construction companies for comparison with competitor performance. The purpose of this study was to survey literatures on the application of DEA methodology and to propose a methodological scheme to measure the performance of construction organizations. Articles on previous studies were surveyed and examined as part of a comprehensive review. The survey revealed that the application of DEA in the construction industry was li mited. Further, the survey indicated that there is a need for the development of a methodological framework on the special goals and subjects of performance measurement, methods of data structure and collection, selection of appropriate DEA models, analysis of results, and post-test. Based on the survey, this study identified and discussed the types of major issues and topics for future studies from a methodological perspective, which could be helpful to researchers interested in using DEA to study performance issues in construction organizations.

• Journal of Intelligence and Information Systems
• /
• v.7 no.1
• /
• pp.63-79
• /
• 2001

Data Envelopment Analysis(DEA), a non-parametric productivity analysis tool, has become an accepted approach for assessing efficiency in a wide range of fields. Despite of its extensive applications and merits, some features of DEA remain bothersome. DEA offers no guideline about to which direction relatively inefficient DMUs improve since a reference set of an inefficient DMU, several efficient DMUs, hardly provides a stepwise path for improving the efficiency of the inefficient DMU. In this paper, we aim to show that DEA can be used to evaluate the efficiency of life insurance companies while overcoming its limitation with the aids of machine learning methods.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.24 no.2
• /
• pp.53-68
• /
• 1999

This paper presents a nonparametric method to test if the mean difference of DEA efficiency estimates between two groups statistically exists. A proposed method employs a bootstrapping approach to generation BCC efficiency estimates through Monte Carlo simulation resampling process. For the purpose of demonstration, we empirically apply the proposed method to the korean bank industry and compare its result with the result by the traditional deterministic DEA method. The nonparametric statistical hypothesis testing procedure in this study, which considers not only stochastic variability of the DEA data, but also random radial deviations off the efficient frontier, serves as a useful tool for dbjectively evaluating whether the mean difference of DEA efficiency estimates between groups is statistically significant.