예금은행 가중평균대출금리가 예금취급기관의 지역별 가계대출에 미치는 영향에 대한 연구

김주일 사단법인 한국신용카드학회 2018 신용카드리뷰 Vol.12 No.3

This study is intended to provide implications by identifying the interconnection between the weighted average lending rate of the deposit bank and the regional household loans of the depository. Methods used in the analysis include Grandeur causal analysis based on the VAR model, impact response function analysis and variance analysis. The main findings of the study are as follows: First, the analysis of the Grandeur causal relationship showed that the weighted average lending rate of the deposit banks affected household loans of regional deposit borrowers, and that the weighted average rate on the basis of the new installment plan was more than the average. Second, the impact response function analysis indicates that the weighted average rate of deposit banks is affecting household loans by region of the deposit receiving institution up to a certain time range, indicating support for the results from the Grandeur causal analysis. Third, as a result of the analysis of variance, household loans of deposit borrowers are affected by a certain percentage of the weighted average interest rate of deposit banks, and the weighted average interest rate of the balance is more influential in regional household loans. Meanwhile, Sejong City, the only 18 regions subject to analysis, was found to be subject to a very small impact on the weighted average interest rate of the depository. Collecting these analyses, the weighted average interest rate of a deposit bank suggests that it is an important factor in determining household loan amount as a precursor to the household loan amount of each region of the depository. The results of this study are expected to help the government and financial sector establish credit policies, such as lending policies and regional lending conditions. This study is meaningful in that it analyzed for the first time the weighted average lending rate of deposit banks on household loans in the region and could continue with various comparative analysis with structural changes. 본 논문은 예금은행 가중평균대출금리와 예금취급기관의 지역별 가계대출 간의 상호 연관성 여 부를 규명하여 시사점을 제시하고자 하였다. 분석에 사용된 방법은 VAR모형에 기초한 그랜저 인 과관계 분석, 충격반응함수 분석 및 분산분해 분석 등이다. 주요 연구결과는 다음과 같다. 첫째, 예금은행의 신규취급액기준 가중평균대출금리는 전국을 포함한 14개 지역별 예금취급기관의 가계 대출에 영향을 미치는 것으로 나타났으나, 4개(강원, 전북, 전남, 세종) 지역에 대하여는 영향을 미 치지 않는 것으로 나타났다. 둘째, 예금은행의 잔액기준 가중평균대출금리는 전국을 포함한 14개 지역별 예금취급기관의 가계대출에 영향을 미치는 것으로 나타났으나, 6개(대전, 강원, 충북, 전북, 전남, 세종) 지역에 대하여는 영향을 미치지 않는 것으로 나타났다. 셋째, 신규취급액기준 가중평 균대출금리보다 잔액기준 가중평균대출금리가 가계대출에 좀 더 영향을 미치는 것으로 나타났다. 한편, 분석대상 18개 지역 중 유일하게 세종시는 예금취급기관의 가중평균대출금리에 매우 작은 영향을 받고 있는 것으로 나타났다. 이와 같은 분석결과를 종합해 보면, 예금은행의 가중평균대출 금리는 예금취급기관의 지역별 가계대출액의 선행지표로서 가계대출액을 결정하는 중요한 요인이 됨을 암시한다. 본 연구의 결과는 정부와 금융권들의 여수신정책과 지역별 대출조건 등 신용정책 을 수립하는데 기여할 것으로 사료된다. 본 연구는 예금은행의 가중평균대출대출금리가 지역별 가 계대출에 미치는 연관성을 최초로 분석하였다는 점에서 의의가 있다고 할 수 있으며, 향후 구조변 화기와의 다양한 비교분석으로 확대될 수 있을 것으로 판단된다.

가중평균에 의한 역계산 어류 성장식추정법 개선 연구

양우성,이재봉,허요원,권대현,최석관,정상덕,안두해 한국수산해양기술학회 2017 수산해양기술연구 Vol.53 No.4

This study aims to suggest the methodology to improve to estimate back-calculated fish growth parameters using weighted average. It is to contribute to correct errors in the calculation of back-calculated growth equation with unequal numbers of sample by age. If the numbers of sample were evenly collected by age, each back-calculated length at age was equal between arithmetic and weighted averages. However, most samples cannot be evenly collected by age in reality because of different catchability by fishing gear and limitation of environment condition. Therefore, the estimation of back-calculated length by weighted average method is essential to calculate growth parameters. There were some published growth equations from back-calculated length using a simple arithmetic average with different numbers of samples by age when searching for back-calculated growth equations from 91 relevant papers. In this study, the process of deriving growth equation was investigated and two different average calculations were applied to a fish growth equation, for example of Acheilognathus signifer. Growth parameters, such as  ∞ ,  and , were estimated from two different back-calculated averages and the growth equations were compared with growth performance index. Based on the correction of back-calculated length using weighted average by age, the changes by female and male were -14.19% and -5.23% for  ∞ , and 59.28% and 18.91% for , respectively. The corrected growth performance index by weighted average improved at 7.05% and 2.46% by female and male, respectively, compared to the arithmetic averages. 수산자원의 성장식 연구는 수산자원 연령구조를 파악하고 자원 상태를 평가하는데 있어 기초적이고 중요한 분야이다. 따라서 가장 적합한 방법과 모델을 선택하여 성장식을 추정하는 과정은 수산자원의 생물학적 특성을 알아내는 의미뿐만 아니라 자원량 산정 및 정도 높은 자원평가를 수행하는데 있어 오차를 줄이고 정확한 평가결과를 도출하기 위해 중요하다. 이에 본 연구에서는 가중평균을 활용하여 역계산체장 및 성장식에 오차를 줄이고 더 정확한 값을 산정함으로써, 과거 수행되어진 연구자료를 비판하기 위한 목적이 아니라, 앞으로의 정확한 수산자원평가 수행을 위해 보다 바람직하게 개선된 값을 입력자료로 제공하기 위한 방법론을 제시하고자, 산술평균이 적용되어 있는 묵납자루의 연령성장(Baek et al., 2002)의 연구결과 토대로 가중평균법을 적용하여 역계산 체장 또는 체중 성장식을 추정하였다. 산술평균이 적용된 연구에 가중평균이 적용되었을 때 성장계수 추정치는 암컷에서 L_(∞ )값은 -14.19%, K 값은 59.28%차이를 나타냈으며, 수컷에서는 L_(∞ )값은-5.23%, K 값은 18.91%로 차이를 보였다. 그리고 t_(0 )값은 암컷에서 46%, 수컷에서 36%로 차이가 확인되었다. 이를 토대로 성장률(GPI값)을 비교했을 때 암컷에서 7.05%, 수컷에서 2.46% 개선된 것을 확인할 수 있었다.

감천유역의 취약도 분석을 위한 MCDM 기법의 비교 연구

김태형,김병현,한건연 한국방재학회 2015 한국방재학회논문집 Vol.15 No.5

의사결정문제를 포함하는 다양한 분야에서 활용되고 있는 MCDM(다기준의사결정법)은 기본가정 및 분석절차에 따라 다양한 기법으로 소개되고 있다. 특히 수자원분야에서도 소유역 또는 행정구역 단위의 홍수취약도를 정량적으로 평가하기 위해 순위화 및 등급화 등을 위한 도구로 다양하게 활용되고 있다. 본 논문에서는 지역단위의 홍수취약도 산정시 적용될 수 있는 다양한 MCDM 기법들의 활용성을 검토하기 위해 Weighted Average, PROMTETHEE, Compromise Programming, VIKOR, 그리고 TOPSIS 기법을 이용하여 감천유역의 홍수취약도를 산정하였다. 기법들의 적용성을 검토하기 위해 홍수취약도 산정시 일반적으로 적용될 수 있는 Weighted Average를 기준으로 하여 각 기법들 간의 취약도 순위에 대한 상관성을 분석하였고, 그 결과 취약도 분석을 위한 MCDM 기법의 활용에 있어 일부 기법을 제외한 대부분의 기법에 대한 적용성을 확인하였다. MCDM, which has been used in many different fields including decision making problem, is being introduced in various ways according to the basic assumption and analysis procedure. Especially in the water resources field, it is used as a raking method in various ways to quantitatively evaluate flood vulnerability of a subbasin and an administrative district. This study measured flood vulnerability of the Gamcheon basin by using Weighted Average, PROMTETHEE, Compromise Programming, VIKOR, and TOPSIS technique in order to examine the capability of various MCDM techniques which can be applied for estimating flood vulnerability of regional units. Based on the Weighted Average which is generally applied for estimating flood vulnerability, this study analyzed the correlation of vulnerability raking between different techniques to examine the technique applicability. Ultimately, it was found that the MCDM technique applications for analyzing vulnerability have applicability on most techniques.

구급방류 음소 비교

김남경 영남대학교 민족문화연구소 2019 민족문화논총 Vol.72 No.-

This study aims to present a statistical processing of phonemic notion in the computerized Korean language history data (Gugeupbang types), propose the distribution of phonemes possessed by Korean language history data in various ways, and provide a tool to find out the variety of meanings in the information of Korean language history data. In this regard, three kinds of computerized Gugeupbang types such as 『Gugeupbang』, 『Gugeupganibang』, 『Eonhae-Gugeupbang』 were statistically quantified and got a special value of initial phonemes, middle phonemes, and final phonemes from individual documents. In the same way, the study was extended to 『Gugeupganibang』 and 『Eonhae-Gugeupbang』 which were changed into a big data and it was obtained an average of the specific value. This study was used the term ‘spectrum analysis’ to visualize the individual phoneme frequency rate among the whole based on the frequency of each phoneme, the frequency rate table, and the three phonemic phonemes. In the comparison of phoneme spectrum, it was calculated the total frequency of initial, middle, and final phonemes, the comparison spectrum of major phoneme frequency rate, the comparison spectrum of rare phoneme frequency rate, the numerical mean spectrum, the weighted average spectrum, and the average correlation spectrum. According to the average correlation of the three documents based on the weighted average, the initial and middle phonemes were appeared 『Gugeupganibang』, 『Eonhae-Gugeupbang』, 『Gugeupbang』 in order and the final phonemes were showed 『Gugeupganibang』, 『Gugeupbang』, 『Eonhae-Gugeupbang』 in order. The results corresponded with the weighted average. Therefore, it can be confirmed that 『Gugeupganibang』 is the highest document of the average agreement in the weighted average. 이 연구는 전산화된 국어사 자료(구급방류)에 나타나는 음소 표기를 대상으로 통계 처리를 하고, 이를 통하여 선명하고 다양한 방법으로 국어사 자료가 지니는 음소의 분포를 제시하여, 국어사 자료가 가진 정보 속에서 다양한 의미를 찾아낼 수 있는 도구를 마련하고자 한 것이다. 이를 위해 전산화한 3종의 구급방류 - 『구급방』, 『구급간이방』, 『언해구급방』을 대상으로, 『구급방』이라는 개별 문헌이 가진 초성·중성·종성 음소를 다양한 방법으로 통계-수치화하여 특정값을 구하고, 이와 같은 방법으로 『구급간이방』과 『언해구급방』으로 확대(빅데이터화)하여 특정값과의 평균을 산출하였다. 이를 나타내기 위한 방법으로 각 문헌별 전체 음소의 빈도와 빈도율 표, 세 문헌 음소 비교표를 바탕으로, ‘스펙트럼 분석’이라는 용어를 사용하여 전체 가운데 개별 음소 빈도율을 시각화하였다. 음소 스펙트럼 비교에는 전체 초성·중성·종성 빈도율 비교 스펙트럼, 주요 음소 빈도율 비교 스펙트럼, 희소 음소 빈도율 비교 스펙트럼, 산술평균 스펙트럼, 가중평균 스펙트럼, 평균 대비 상관도 스펙트럼 등을 산출하였다. 세 문헌의 평균 대비 상관도 비교에서는 가중평균을 기준으로, 초성과 중성에서는 『구급간이방』 > 『언해구급방』 > 『구급방』의 순서로, 종성에서는 『구급간이방』 > 『구급방』 > 『언해구급방』의 순서로 가중평균과의 일치도를 보였다. 그러므로 가중평균에서는 『구급간이방』이 평균 대비 일치도가 가장 높은 문헌임을 확인할 수 있었다.

The average trapping time for a weight-dependent walk on a weighted hierarchical graph

Wu Bo,Cao Fang,Chen Yun 한국물리학회 2021 THE JOURNAL OF THE KOREAN PHYSICAL SOCIETY Vol.78 No.12

A hierarchical graph is a kind of self-similar network, which is widely discussed and has a wide range of applications. In this paper, we introduce a class of weighted hierarchical graphs H (n, k), which is constructed using the hierarchical product of complete graphs. The weighted hierarchical graph depends on two parameters: the weighted factor r (0 < r ≤ 1) and the number k (k ≥ 3) of nodes of the complete graph K k . In addition, we also discuss the trapping problem and the average trapping time (ATT) for a weight-dependent walk on a weighted hierarchical graph. Based on the special structure of the weighted graph, we derived the exact expression of the ATT for a weighted hierarchical graph. The results show that the ATT obeys a power-law function with the exponent ln(kr) lnk < 1 , which is related to the factors k and r and indicate that for kr = k(r = 1) and 1 < kr < k , the ATT has a linear and sublinear relationship with the network order, respectively. Therefore, the smaller kr is, the more efficient the trapping process of the weighted hierarchical graph is. Compared with the unweighted hierarchical graph, the weighted hierarchical graph has a higher efficiency in the trapping process.

열수처리와 투석이 alginates의 평균분자량의 측정에 미치는 영향

유병진 ( Yu Byeong Jin ),임영선 ( Im Yeong Seon ),류홍수 ( Lyu Hong Su ) 한국수산과학회 2004 한국수산과학회지 Vol.37 No.1

N/A In preparing sodium alginates from sea tangle (Laminaria japonica) powder using the Mexican process, alkaline extraction, conversion to alginic acid and reversion to sodium alginates were used to increase purity. The effect of hot water treatment and dialysis on measuring the average molecular weight of sodium alginates were investigated. Intrinsic viscosity and average molecular weight of sodium alginates after dialysis were higher than those before dialysis. Average molecular weight of sodium alginates treated with hot water was higher than that without. Hot treatment has little effect on the ash content of sodium alginates. Ash content of sodium alginates before dialysis were 27-30%; those after dialysis were 10%. After dialysis, Na content was highest (89-91%), K was 11-12%, Ca was 1.9%, and Mg was 0.05%. Ash content of alginates had little effect on average molecular weight. SAV (slope of apparent viscosity) of alginates solution after dialysis showed higher values than before. SAV of the alginates with hot water treatment were higher than without treatment.

Correlation between reproductive performance and sow body weight change during gestation

Sang Hun Ha,최요한,문준영,Serin Park,Elick Kinara,박현주,홍준선,Yong-Min Kim,Jin-Soo Kim 한국축산학회 2024 한국축산학회지 Vol.66 No.3

This study investigated the correlation between piglet performance and sow body weight change (BWC) during two gestational periods: 35–70, 70–105, and 35–105 days. A cohort of 70 sows was evaluated for BWC, backfat thickness change (BFC), caliper score change (CALC), feed intake, and weaning-to-estrus interval (WEI). The collected data were then analyzed according to the two specified periods. Our findings highlighted that piglet birth weight, weaning weight, and average daily weight gain (ADG) correlated with sow body characteristics, including BFC and CALC. The strongest correlation was observed with BWC. Piglet mortality was intimately associated with BFC. Piglet birth weight, weaning weight, and ADG showed a positive correlation with sow BWC, particularly during the 35–70 day period. Furthermore, sows displaying a higher BWC during the 70–105 day period, and also exhibiting a higher BW gain from 35-70 days, registered greater piglet weight gains and higher weaning weights. These trends became more apparent as the sow’s BWC increased during the 70–105 day period. Piglet mortality increased when the sow exhibited a lower BWC during both the 35–70 and 70–105 day periods. No significant observations were found concerning the number of stillborn piglets, live-born piglets, or weaned piglets, and no interaction effects were detected between these periods. In conclusion, our findings underscore the significance of sow BWC during the early stages of gestation (d 35–70) for enhancing piglet performance from birth to weaning.

최대우도 가중평균 신경망을 이용한 객체 위치 추적

박선배,유도식 한국항행학회 2023 韓國航行學會論文誌 Vol.27 No.1

Object tracking is being studied with various techniques such as Kalman filter and Luenberger tracker. Even in situations, such as the one in which the system model is not well specified, to which existing signal processing techniques are not successfully applicable, it is possible to design artificial neural networks to track objects. In this paper, we propose an artificial neural network, which we call 'maximum-likelihood weighted-average neural network', to continuously track unpredictably moving objects. This neural network does not directly estimate the locations of an object but obtains location estimates by making weighted average combining various results of maximum likelihood tracking with different data lengths. We compare the performance of the proposed system with those of Kalman filter and maximum likelihood object trackers and show that the proposed scheme exhibits excellent performance well adapting the change of object moving characteristics. 객체 위치추적은 칼만필터나 루엔버거 추적기와 같은 다양한 기법을 통해 연구되고 있는데, 시스템 모델이 명확하게 규정되지 않는 경우와 같이 기존의 신호처리 기법을 성공적으로 적용하기 어려운 상황에서도 객체의 위치를 추적해 낼 수 있는 인공신경망을 설계하는 것이 가능하다. 본 논문에서는, 불규칙하게 운동 상태가 변화하는 객체의 위치를 지속적으로 추적하기 위해, 다양한 길이의 최대우도기법 객체위치 추정깂들을 도출한 뒤 신경망을 통해 이들을 적응적으로 가중평균하는 방법을 사용하는 ‘최대우도 가중평균 신경망’을 제안한다. 해당 신경망은 객체의 위치를 직접 추정하지 않고 데이터 길이가 다른 다양한 최대우도기법 추적 결과들을 가중평균하여 위치 추정을 수행한다. 우리는 제안하는 시스템의 추적성능을 칼만필터 및 최대우도기법들과 비교하여, 제안하는 기법이 물체의 움직임 특성에 잘 대처하여 우수한 성능을 나타내 줌을 보인다.

Fractional Group Identification

Wonki Jo Cho,Chang Woo Park 한국재정학회(구 한국재정·공공경제학회) 2018 한국재정학회 학술대회 논문집 Vol.2018 No.-

We study group identification problems, the objective of which is to classify agents into groups based on individual opinions. Our point of departure from the literature is to allow membership to be fractional, to qualify the extent of belonging. Examining implications of independence of irrelevant opinions, we identify and characterize four nested families of rules. The four families include the weighted-average rules, which are obtained by taking a weighted average of all entries of a problem, and the fractional consent rules, which adapt the consent rules from the binary model to our multinary setup, balancing two principles in group identification, namely liberalism and social consent. Existing characteri-zations of the one-vote rules, the consent rules, and the liberal rule follow from ours.

퍼지 소속도 함수와 가중치 평균을 이용한 지도 학습 기반 분류기 설계

우영운 한국정보통신학회 2021 한국정보통신학회논문지 Vol.25 No.4

'In this paper, to propose a classifier based on supervised learning, three types of fuzzy membership functions that determine the membership of each feature of classification data are proposed. In addition, the possibility of improving the classifier performance was suggested by using the average value calculation method used in the process of deriving the classification result using the average value of the membership degrees for each feature, not by using a simple arithmetic average, but by using a weighted average using various weights. To experiment with the proposed methods, three standard data sets were used: Iris, Ecoli, and Yeast. As a result of the experiment, it was confirmed that evenly excellent classification performance can be obtained for data sets of different characteristics. It was confirmed that better classification performance is possible through improvement of fuzzy membership functions and the weighted average methods. 본 논문에서는 지도 학습 기반의 분류기 제안을 위해, 분류 데이터의 각 특징별 소속도를 결정하는 3가지 종류의 퍼지 소속도 함수를 제안하였다. 또한 각 특징별 소속도들의 평균값을 이용하여 분류 결과를 도출하는 과정에 사용되는 평균값 산출 기법을 단순 산술평균이 아닌 다양한 가중치를 활용한 가중치 평균을 이용함으로써 분류기 성능을 향상시킬 수 있는 가능성을 제시하였다. 제안한 기법들의 실험을 위해 Iris, Ecoli, Yeast의 3가지 표준 데이터 세트를 사용하였다. 실험 결과, 서로 다른 특성의 데이터 세트들에 대해서도 고르게 우수한 분류 성능이 얻어질 수 있음을 확인하였고, 기존에 발표된 다른 기법들에 의한 해당 데이터 세트들의 분류 성능과 비교했을 때, 퍼지 소속도 함수의 개선과 가중치 평균 기법의 개선을 통해 더욱 우수한 분류 성능이 가능함을 확인할 수 있었다.