• 한국지구과학회:학술대회논문집
• /
• 2010.04a
• /
• pp.80-80
• /
• 2010

본 연구에서는 식물의 개화시기가 기온에 민감하게 반응한다는 것에 근거하여 54개의 기상관측지점을 선택하고 1988년부터 2008년까지 21년간 벚꽃 개화일의 시계열적 변화를 분석하였고 이들과 3월 평균기온과의 상관관계를 조사하였다. 그리고 엘니뇨와 라니냐 시기에 기온과 벚꽃개화일의 변화 경향과 도시화에 의한 기온변화가 벚꽃 개화일에 미치는 영향을 대도시와 교외도시와의 비교를 통해 알아보았다. 또한 경험직교함수(EOF) 분석을 통하여 벚꽃개화일의 시 공간 분포를 알아보았다. 그 결과는 다음과 같다. 첫 번째, 3월 평균기온 분석을 통해 1988-2008년 사이에 한반도의 기온은 평균적으로 $0.13^{\circ}C$ 증가하였음을 확인할 수 있다. 그리고 평균적으로 남해안이 동해안과 서해안보다 평균기온이 높게 나타났으며, 서해안이 가장 낮은 평균기온을 보였다. 그리고 내륙지역의 경우 위도에 따라 평균기온이 달랐으며, 위도가 낮아질수록 평균기온이 크게 나타났다. 하지만 모든 지점에서 유의하지는 않았다. 두 번째, 벚꽃개화일의 변화를 보면, 개화일은 시간이 지날수록 빨라지는 경향을 보였다. 그리고 벚꽃개화일의 분포를 보면 위도가 낮은 지역이 개화일이 빨랐으며, 해안지역 중에서는 남해안, 동해안, 서해안 순이었다. 세 번째, 2월, 3월, 4월의 평균기온과 벚꽃 개화일의 상관관계 분석결과 3월 평균기온이 벚꽃개화일과 상관관계가 가장 높았다. 네 번째, 엘니뇨가 발생한 해에는 한반도의 기온이 상승하는 경향을 보였으며, 동시에 벚꽃개화일도 빨라졌다. 다섯 번째, 대도시가 교외도시에 비해서 기온상승이 높았으며, 벚꽃개화일도 빠르게 나타났다. 마지막으로, 경험직교함수(EOF) 분석 결과 3월 평균기온과 벚꽃 개화일의 1mode 시계열 패턴이 비슷하다는 것을 확인하였다. 또한, 1996년을 기준으로 이전에는 개화일이 느려지고, 이후에는 빨라지는 패턴을 보였다.

• The Korean Journal of Financial Studies
• /
• v.10 no.1
• /
• pp.145-170
• /
• 2004

Roll의 비판 이후 실행된 많은 국내외 연구결과 CAPM으로 설명이 되지 않는 이례 현상(Anomaly)들이 발견되고 있다. 이례 현상들은 다 요인 모형(multi-factor model)과 같은 추가 위험 요인이론, 표본차이이론, 과잉반응 및 특성이론들로 설명되고 있고 이러한 이례 현상들은 재무관리의 지속적인 관심사인 미래의 주가수익률 예측과 밀접한 관계에 있다. 본 연구에서는 이례 현상들이 주가수익률에 미치는 영향을 알아보기 써하여 Haugen and Baker(1996)의 다 요인 및 수익률 추정 방법론을 국내 증권시장에 적용한 다 요인 모형과 $\beta$, 기업규모, PBR, 과거 1년 주가 수익률에 의한 단일 요인 모형을 이용하여 개별 기업의 포트폴리오 구성기준을 결정하고 이 기준에 의거하여 월별로 편입 주식들을 재조정한 포트폴리오들의 년간 누적 실제수익률 예측력을 비교 분석한 결과 다음과 같은 결과를 얻었다. 첫째, 다 요인모형의 경우 기대수익률이 높은 주식으로 구성된 포트폴리오가 기대수익률이 낮은 주식으로 구성된 포트폴리오보다 실제 년간 수익률이 높게 나타난 반면, $\beta$, 기업규모, PBR, 과거 1년 주가 수익률의 요인에 의한 단일 모형을 적용한 포트폴리오는 이들 순위와 실제 수익률간에는 상관성이 높지 않게 나타나 다요인 모형이 주가 수익률 예측력에 있어서 단일요인 모형보다 우수한 것으로 판단된다. 단일모형 중에서는 PBR을 이용한 포트폴리오가 $\beta$ 단일모형보다 좋은 주가수익률 예측력을 보여 주었다. 둘째, 주가 수익률을 결정하는 유의성있는 요인들은 당기순이익의 증감, 당해연도의 당기순이익의 분포, 자산증가율, 매매 유동성, 매출액 변동, 거래량 추세, 기업크기(시가총액), 과거 1개월간의 주가수익률, 자기자본증가율등으로 나타났다.

• Journal of the Economic Geographical Society of Korea
• /
• v.17 no.3
• /
• pp.482-501
• /
• 2014

Most textbooks of urban geography explain that land value drops down with distance from the Peak Land Value Intersection of urban center. However, There is little rigorous empirical work supporting this belief. Using the publicly notified individual land price data of 2013, this paper examines the structure of land value in urban center with a focus on spatial distribution of land value by land use. As an analytic result, the structure of land price varies between land uses, significantly being different from the classical model of land value in urban space. In residential lands and mixed-use lands for residence and commerce, land price gradient as a function of distance from PLVI is positive and a crater phenomenon even is identified in spatial distribution of land value. The classical model coincides only with land for commerce and office. Nonetheless, predict of land uses by land value function fairly corresponds to Alonso's model.

• Journal of Korean Society of Forest Science
• /
• v.102 no.2
• /
• pp.292-299
• /
• 2013

The purpose of this study is to identify the regional specialization of Fruit trees in Korea and to provide direction for the activation. As a result of the analysis, In the case of chestnut, Gongjun is adjacent to Cheongyang and Buyeo. These region are classified as deeply specialized regions. Walnut trees are grown in large amounts in Gimcheon-si, Yeongdong-gun and Muju-gun, which are all regions classified as deeply specialized regions. Jujube trees are grown mostly Gyeongsan-si, Gunwi-gun, Cheongdo-gun and Yeongcheonsi are geographically close to each other. The production quantity of these areas also showed steady growth. These region are classified as deeply specialized regions. Astringent persimmons are Sangju-si and Yeoungdonggun are close to each other. Other regions are scattered over the country.

• Journal of the Korean Geographical Society
• /
• v.48 no.3
• /
• pp.366-386
• /
• 2013

This paper analyzes the drought and restriction on water supply in Taebaek City during the winter season in 2008 using Actor-Network Theory. Actor-Network Theory emphasizes and brings into view the role and act of non-human actors as well as human actors in various environmental issues. The fact that only Taebaek experienced restriction on water supply for 88 days although the winter season drought in 2008 affected the whole nation, requires a synthetic analysis of both human and non-human actors and their relationships and networks embedded in Taebaek City at that time. This paper shows that both human and non-human actors including Taebaek City Hall, Korea Water Resource Corporation, Taebaek citizen, the water supply facilities, Gwangdongdam, obsolete water pipes, the topography of Taebaek, soil, the change of industry, and population interact one another transforming the geography of water in Taebaek. This study helps to understand the complex processes related to drought disasters at a specific local scale and to provide appropriate measures to drought.

• The KIPS Transactions:PartB
• /
• v.15B no.6
• /
• pp.603-608
• /
• 2008

In this paper an approach to building fuzzy model to predict daily open, close, high, and low stock prices is presented. One of prior problems in building a stock prediction model is to select most effective indicators for the stock prediction. The problem is overcome by the selection of information used in the analysis of stick-chart as the input variables of our fuzzy model. The fuzzy rules have the premise and the consequent, in which they are composed of trapezoidal membership functions, and nonlinear equations, respectively. DE(Differential Evolution) searches optimal fuzzy rules through an evolutionary process. To evaluate the effectiveness of the proposed approach numerical example is considered. The fuzzy models to predict open, high, low, and close prices of KOSPI(KOrea composite Stock Price Index) on a daily basis are built, and their performances are demonstrated and compared with those of neural network.

• Spatial Information Research
• /
• v.19 no.1
• /
• pp.1-11
• /
• 2011

Recently urban area has suffered from frequent flood event by local heavy rain. This study performed flood tests for the Jungnang river using HEC-RAS model. Based on 1m LiDAR data, river geometry data were produced using HEC-GeoRAS. For 100-year frequency flood, 200-year frequency flood, and PMF, flooding areas were estimated. Ten sub-zones of the entire flooding area were identified based on the nearest refugees and used to analyze evacuation paths to the refugees. The results showed that approximately 70% of flooded area were residential, commercial, and transportation areas so that much loss of life and property could be possible. Path analysis showed that the shortest path distances to refugees were about 1000m average. Evacuation warning given at a proper period could minimize loss of life and property. This study provides the guideline for flood evacuation plan in urban area.

• Asian Journal of Turfgrass Science
• /
• v.17 no.1
• /
• pp.45-51
• /
• 2003

The purpose of this study was to find out how different types of land use around green belts affect temperature decrease effects. For this, temperatures within and outside of green belts were measured. Based on the collected data, the study analyzed the land cover status and temperatures within green belts, temperature decrease effects and the range of effects around green belts, and the correlation between trees and temperature decrease effects by way of regression analysis. As a result, areas of the high temperature within green belts were formed on paved surfaces, whereas areas of low temperature were formed around forests or water surfaces. In addition, deviation was bigger in the highest temperature than the lowest one for areas of Leeward around green belts, but in general, there was a tendency that temperature became low near to green belts. As for the relation between land cover rate and temperature, what was effective to temperature decrease included forests, pasture and water surfaces. On the other hand, the effects of temperature decrease varied depending on increase or decrease of land cover rates. As for the influence of the different land use types around green belts on temperature decrease effects, the Shakuzi Park showed relatively stronger effects than the Ageomaruyama Park.

• Journal of the Korean Society for Railway
• /
• v.20 no.5
• /
• pp.692-702
• /
• 2017

Recently, there has been a growing interest in using smart card data. However, there are restrictions on the utilization of data in many areas outside the Seoul metropolitan area because the data does not contain alighting information. This paper presents a methodology for estimating alighting stops of smart card data. Estimation results were verified by smart card data from Seoul and Gwangju. The estimation rates were 78.2% and 81.6% in Seoul and Gwangju, respectively. The matching accuracy was 54.2% and 33.4%, respectively. However, if up to two stops of error are allowed, the accuracy values were 93.6% and 94.0%, respectively. We also discussed changes in estimation results due to adjusting the allowable walking distance, which is a key parameter of trip chaining methods. As the allowable walking distance increases, the estimation rate increases, while the accuracy decreases, and it is found that the estimation results change by around 500m.

• Korean Journal of Plant Resources
• /
• v.30 no.4
• /
• pp.399-409
• /
• 2017

The rapid increase of naturalized plants causes disturbance of the ecosystem. The purpose of this study is to identify the reality of naturalized flora in Jeolla-do. A total of 830 sites were studied in 2016. These naturalized plants consisted of 189 taxa in total, belonging to 38 families, 116 genera, 181 species, 2 forms and 6 varieties. According to the results of the analysis based on place of origin, 75 taxa were from Europe, and 59 taxa were from North America. According to the analysis of the ecological characteristics, compositae had the highest diversity among 31 families, with 44 taxa, followed by Gramineae, with 24 taxa, and Leguminosae, with 10 taxa. Naju-si had the highest diversity among 36 counties, with 77 taxa. This study identified 37 (Jeollanam-do) and 30 taxa (Jeollabuk-do) not listed in the preceding literature in Jeolla-do. We found a lots of differences between flora's databases of previous studies resulting from the naturalized plant definition.