Earticle

The purpose of this study is to analyze the effect of intervention variables which may affect the air travel demand for Jeju domestic flights and to anticipate the air travel demand for Jeju domestic flights. The air travel demand forecasts for Jeju domestic flights are conducted through ARIMA-Intervention Model selecting five intervention variables such as 2002 World Cup games, SARS, novel swine-origin influenza A, Yeonpyeongdo bombardment and Japan big earthquake. The result revealed that the risk factor such as the threat of war that is a negative intervention incident and occurred in Korea has the negative impact on the air travel demand due to the response of risk aversion by users. However, when local natural disasters (earthquakes, etc) occurring in neighboring courtiers and global outbreak of an epidemic gave the negligible impact to Korea, negative intervention incident would have a positive impact on air travel demand as a response to find alternative due to rational expectation of air travel customers. Also we realize that a mega-event such as the 2002 Korea-Japan World Cup games reduced the air travel demand in a short-term period unlike the perception in which it will increase the air travel demand and travel demands in the corresponding area.

본 논문은 여러 개의 독립적인 시계열로 구성된 시계열 패널 자료를 이용하여 비선형 모형인 GRCA모형과 신경망을 이용하여 예측값을 구하여 서로 비교 분석하고자 한다. 먼저 GRCA모형에 대하여 연구하고 신경망의 구조와 예측값을 구하기 위한 여러 가지 변환함수를 유도한다. 단기 예측에서는 신경망 방법의 예측값이 더 좋았고, 장기예측에서는 비선형모형을 이용한 예측값이 더 좋은 것으로 나타났다.

정규식생지수(NDVI)는 식생자원을 모니터링할 수 있도록 설계된 식생지수(VI-Vegetation Index) 중 하나로 여러 응용 분야에서 가장 많이 사용되고 있는 지수이다. 산림 분야에서도 NDVI가 많이 활용되고 있는데 본 논문에서는 산림 변화 모니터링을 위해 MODIS NDVI를 활용하는 방법론이 연구되었다. 특정 시점을 기준으로 NDVI 값을 비교 및 분류하여 변화를 탐지하는 방법은 기계나 기상상태의 영향으로 자료의 정확성이 떨어질 수 있고 장기적인 변화를 탐지하는데도 어려움이 있다. 이러한 점을 고려하여 본 논문에서는 하모닉 모형을 이용하여 NDVI 시계열 자료를 통해 NDVI 패턴을 고려하는 방법론을 제시하였다. 먼저 하모닉 모형을 적용하여 미관측 자료나 자료의 오류를 보정한 NDVI 시계열 자료를 재구축하고 추정된 하모닉 요소의 모수를 기준으로 장기적 패턴을 통해 식생의 변화를 모니터링할 수 있다. 제안된 방법은 한반도 지역의 2009년 8월 21일부터 2011년 9월 6일까지 총 49개의 MODIS NDVI 시계열 자료에 적용하여 모형의 유용성을 입증하였다.

Normalized Difference Vegetation Index (NDVI) has been used to measure and monitor plant growth, vegetation cover, and biomass from multispectral satellite data. It is also a valuable index in forest applications, providing forest resource information. In this research, an approach for monitoring forest change using MODIS NDVI time series data is explored. NDVI difference-based approaches for a specific point in time have possible accuracy problems and are lacking in monitoring long-term forest cover change. It means that a multi-time NDVI pattern change needs to be considered. In this study, an efficient methodology to consider long-term NDVI pattern is suggested using a harmonic model. The suggested method reconstructs MODIS NDVI time series data through application of the harmonic model, which corrects missing and erroneous data. Then NDVI pattern is analyzed based on estimated values of the harmonic model. The suggested method was applied to 49 NDVI time series data from Aug. 21, 2009 to Sep. 6, 2011 and its usefulness was shown through an experiment.

본 논문에서는 왜곡 제거 시계열 서브시퀀스 매칭에서 인덱스를 효율적으로 구성하는 방법을 다룬다. 기존의 왜곡 제거 시계열 서브시퀀스 매칭에서는 인덱스 구축에 매우 많은 시간이 걸리는데, 이는 왜곡 제거의 모든 가능한 경우를 고려하기 위해 너무 많은 윈도우가 생성되기 때문이다. 실제로 길이 30 만의 시계열인 경우에도 인덱스 구축을 위해 약 100분의 많은 시간이 걸려, 대용량 시계열 데이터에 대해서는 인덱스 구축이 매우 어려운 단점이 있다. 본 논문에서는 기존 인덱스 구축 과정을 단계별로 정형적으로 분석한 후, 각 단계별 성능 극대화 방법을 제안한다. 특히, 동적 프로그래밍 기법을 이용하여 PAA-버킷 및 DF-버킷(distortion-free bucket)의 개념을 제안하는데, 이는 반복되는 계산 결과를 저장-후-재사용(store-and-reuse)하는 기법으로, 이를 사용하여 기존 방법에 비해 효율적인 인덱스 구축이 가능하다. 본 논문에서는 복잡도 분석 및 실험 평가를 통해 제안한 방법의 우수성을 입증한다.

In this paper we present an efficient approach of constructing a multidimensional index in distortion-free time-series subsequence matching. Index construction of previous distortion-free subsequence matching algorithms is a very time-consuming process since it generates a huge number of windows to consider all possible positions and all possible query lengths. According to the real experiment, the index construction time reaches approximately 100 minutes for a time-series of length 300K, and this means that the index construction itself is very difficult for very large time-series databases. To solve this problem, in this paper we first thoroughly analyze the index construction steps, then discuss how to improve the performance of each step, and finally propose two advanced algorithms of efficiently constructing an index. In particular, by exploiting dynamic programming techniques, we present the concepts of PAA-bucket and DF(distortion-free)-bucket, which store-and-reuse the intermediate results repeatedly computed. Through the store-and-reuse technique, the proposed algorithms construct a multidimensional index very faster than the previous algorithm. Through analytical and empirical evaluations, we showcased the superiority of the proposed algorithms.

사용자의 관심사를 시간의 흐름을 고려하지 않고 정적으로 모델링할 경우, 사용자 관심사의 변화 추세(증가/감소 등)를 비롯한 개인 성향 분석에 유용한 여러 정보들이 누락될 수 있다. 본 연구에서는 이와 같은 사용자 성향 정보의 누락을 최소화하기 위하여 사용자 별 시계열 관심 패턴 모델링 방법 및 기존 관심사에 근거하여 새로운 관심사 및 관심 패턴을 추정하는 신규 관심사 추정 방법 등을 제안한다. 제안방법의 타당성은 다음 뉴스 댓글 데이터를 통한 실험으로 검증되었다.

Modeling user's interests without considering the stream of time can ignore some important information for personal preference analysis. In this paper, we propose a method for modeling user representative temporal interest pattern to minimize ignoring important information for personal preference and a method for estimating new interests from existing interests. The validity of the proposed method is justified by experiments with data from 'Daum' news Comments.

본 논문에서는 연속형-GARCH 시계열 자료인 금융 시계열 자료에 대해서 클리핑(clipping)을 통해 얻은 이항(binary) 범주형 시계열을 분석하고 응용하는 방안에 대해 연구하고 있다. 모수추정 방법을 소개하고 있으며 이를 이용하여 이분산 시계열과 연관된 확률을 추정하는 방법을 예시하였다.

This short article is concerned with a categorical time series obtained after clipping a heteroscedastic GARCH process. Estimation methods are discussed for the model parameters appearing both in the original process and in the resulting binary time series from a clipping (cf. Zhen and Basawa, 2009). Assuming AR-GARCH model for heteroscedastic time series, three data sets from Korean stock market are analyzed and illustrated with applications to calculating certain probabilities associated with the AR-GARCH process.

본 연구는 최근에 활발히 연구가 진행 중인 항공수요 예측을 위하여 계절형 다변량 시계열 모형을 기반으로 하고 다른 모형과의 비교를 RMSE(Root Mean Square Error)를 기준으로 비교한 것이다. 여기서 싱가폴 국제항공유가, 수출액을 추가하여 예측성능을 좋게 하고자 한다.

Forecasting for air demand such as international passengers and freight has been one of the main interests for air industries. This research has mainly focus on the comparison of the performances of the multivariate time series models. In this paper, we used real data such as exchange rates, oil prices and export amounts to predict the future demand on international passenger and freight.

Objectifying claims filed during the warranty period, analyzing the current circumstances and improving on the problem in question is an activity worth doing that could reduce the likelihood of claims to occur, cut down on the costs, and enhance the corporate image of the manufacturer. Existing analyses of claims are confronted with two problems. First, you can't precisely assess the risks of claims involved by means of the value of claims per 100 products alone. Second, even in a normal state, the existing approach fails to capture the probabilistic conflicts that escape the upper control limit of claims, thus leading to wrong control activities. To solve the first problem, this paper proposed that a time series detection concept where the claim rate is monitored based on the date when problems are processed and a hazard function for expression of the claim rate be utilized. For the second problem, this paper designed a model whereby to define a normal state by making use of PID (Proportion, Integral, Differential) and infer by way of a fuzzy concept. This paper confirmed the validity and applicability of the proposed approach by applying methods suggested in the actual past data of warranty claims of a large-scaled automotive firm, unlike hypothetical simulation data, in order to apply them directly in industrial job sites, as well as making theoretical suggestions for analysis of claims.

바람길 도입을 통한 도시열섬현상의 저감 방안에 관한 선행연구들을 바탕으로 본 연구에서는 도시화과정을 거치면서 발생하는 토지피복의 변화가 도시공간의 찬공기 생성 및 유동에 어떠한 변화를 일으키는지 분석하였다. 연구 대상지로는 우리나라의 대표적인 극서(劇暑)도시이며 1980년대 이후 주변 지역을 편입하여 도시가 확장된 대구시 내에서도 특히 지표온도의 변화가 크고 1988년 신설되어 급격한 인구 증가가 이루어졌던 달서구를 대상으로 연구를 실시하였다. 시간적 범위로는 1987년, 1997년, 2007년의 여름철 위성영상을 이용하여 20년간의 도시미기후 변화를 연구하였다. 연구결과 자연 토지피복의 감소와 인공 토지피복의 증가는 찬공기 생성 및 유동에 불리하여 도심지에서의 찬공기의 양과 높이가 현저히 낮아졌다. 이는 찬공기 생성 및 유동기능이 토지피복에 의해 영향을 받는 것으로 보였다. 지속적인 찬공기 생성 및 유동의 확보를 위해서는 도시를 둘러싼 주변지역의 공간적 특성을 분석하여 도시개발이나 도시재생을 실시해야 하며, 이를 통해 쾌적하고 건강한 도시조성이 될 수 있을 것으로 사료된다.

Based on previous research on ways of reducing an urban heat island phenomenon via an introduction of wind corridors, I conducted this study to see what influence a change in land cover arising of or going through urbanization has on wind corridors of urban space. As a target place, I chose Daegu city where is a representative extreme heat place in Korea and has been also largely expanded in size by incorporating its neighboring areas since the 1980s, expecially Dalseo District whose surface temperature gap is large. The population of Dalseo District has been sharply increased since its creation as a new administrative district in 1988. I studied on the urban microclimate change for a 20-year period by using satellite images on summer months in 1987, 1997 and 2007 in time frames. The finding of this study found that a reduction of natural land cover and an increase of artificial land cover serves as a disadvantageous factor for cold air creation and flowing and strikingly lowers the amount and height of cold air in the downtown area. It seemed that the cold air creation and flowing functions are influenced by land cover. In order to steadily create cold air and secure its flowing, it is thought that urban development or urban regeneration should be implemented by analysing the characteristics of the space surrounding the city. By doing so, a pleasant and healthy city could be formed.

The demands of residents can be predicted by analyzing the housing viewpoint of residents. And also, the housing viewpoint can change according to various environmental changes, so it's important for us to understand the trend of change. Therefore, the housing viewpoints of university students who will be subjective residents of houses in future were analyzed by observing change due to trend of time. Target for this study is female university students in 20s. A portion of previously presented material (Korean Home Economics Association 37,1, 67-78) was used as data for 1998 and a questionnaire with 171 female university students was conducted on December 2008 for data for 2008. The following result was drawn from examination of change in the housing viewpoint due to trend of time by comparing the housing viewpoint of university students between 1998 and 2008. First, importance of <Function as a place for happiness and harmony of family> was decreased and importance of <Function as a place for health and rest of family> was increased in functions of residence. Second, regardless of the time passage, the most students thought <A house is a place for happiness of family> and <A place for physical and mental rest> as their opinions on housing. And also, the most students considered a living room as the most important space inside of a house regardless of the time passage and the level of considering a living room as the most important space was more increased. Third, the level of consideration got higher with time in the selection of houses. <internal facilities>, <safety>, <lighting and ventilation>, <number of rooms>, <convenience facility>, <area of room> and <Security of private life> had higher level of consideration regardless of time passage. Items of which level of consideration increased with time are <safety>, <number of rooms>, <convenience facility>, <internal space>, <easy sales and purchase>, <reputation of village> and <investment prospect>, and items of which level of consideration decreased are <exterior design> and <silence and secludedness>. Fourth, it can be understood that the level of satisfaction in housing got higher with time. And it can be said that <area of room>, <silence and secludedness>, <scenic green area>, <easy sales and purchase> and <investment prospect> were not increased in the level of satisfaction in housing compared with 10 years ago. Fifth, regardless of time passage, the level of satisfaction in housing was lower than the level of consideration in the event of selection of houses. As <internal space>, <safety>, <exterior design> and <area of room> had lower levels of satisfaction in the event of selection of houses compared with the levels of consideration, they have higher possibility of dissatisfaction in housing.

Housing distribution rate in our country has been continuously increased with economic growth, but residential plans satisfying the demands of residents are still not perfect yet. The demands of residents can be predicted by analyzing the housing viewpoint of residents. And also, the housing viewpoint can change according to various environmental changes, so it's important for us to understand the trend of change. Therefore, the housing viewpoints of university students who will be subjective residents of houses in future were analyzed by observing change due to trend of time. Target for this study is female university students in 20s. A portion of previously presented material (Korean Home Economics Association 37,1, 67-78) was used as data for 1998 and a questionnaire with 171 female university students was conducted on December 2008 for data for 2008. The following result was drawn from examination of change in the housing viewpoint due to trend of time by comparing the housing viewpoint of university students between 1998 and 2008. First, importance of <Function as a place for happiness and harmony of family> was decreased and importance of <Function as a place for health and rest of family> was increased in functions of residence. Second, Most students thought <A house is a place for happiness of family> and <A place for physical and mental rest> regardless of the time passage as their opinions on housing. And also, most students considered a living room as the most important space inside of a house regardless of the time passage and the ratio of considering a living room as the most important space was more increased.

본 논문에서는 주식시장에서 거래되는 다수의 주식거래종목들을 몇 개의 그룹으로 군집화하는 주제를 연구한다. 시간에 관계없이 분산이 일정한 ARMA모형과 다르게, 주가, 환율 등의 금융시계열자료에서는 조건부 이분산성을 따르게 된다. 또한, 많은 사람들이 금융시계열자료에서 관심을 갖는 것은 바로 이 변동성이다. 그러므로, 이 연구에서는 조건부 이분산성을 모형화하기에 적합하다고 알려진 일반화 조건부 이분산성 자기회귀모형에 초점을 맞춘다. 먼저 두 개의 주식종목들 사이에 변동성(volatility)의 유사성 그리고 구조의 유사성을 재는 거리를 정의하고, 모의실험을 수행한다. 실증자료로 최근 3년 동안 관찰된 국내 11개 주가의 수익률을 변동성과 구조에 따라 군집화한다.

In this study, we considered the clustering analysis for stock return traded in the stock market. Most of financial time-series data, for instance, stock price and exchange rate have conditional heterogeneous variability depending on time, and, hence, are not properly applied to the autoregressive moving-average(ARMA) model with assumption of constant variance. Moreover, the variability is font and center for stock investors as well as academic researchers. So, this paper focuses on the generalized autoregressive conditional heteroscedastic(GARCH) model which is known as a solution for capturing the conditional variance(or volatility). We define the metrics for similarity of unconditional volatility and for homogeneity of model structure, and, then, evaluate the performances of the metrics. In real application, we do clustering analysis in terms of volatility and structure with stock return of the 11 Korean companies measured for the latest three years.

Recently, causality analysis of source time series extracted from EEG or MEG signals is becoming of great importance in human brain mapping studies and noninvasive diagnosis of various brain diseases. Two approaches have been widely used for the analyses: one is independent component analysis (ICA), and the other is multiple signal classification (MUSIC). To the best of our knowledge, however, any comparison studies to reveal the difference of the two approaches have not been reported. In the present study, we compared the performance of the two different techniques, ICA and MUSIC, especially focusing on how accurately they can estimate and separate various brain electrical signals such as linear, nonlinear, and chaotic signals without a priori knowledge. Results of the realistic simulation studies, adopting directed transfer function (DTF) and Granger causality (GC) as measures of the accurate extraction of source time series, demonstrated that the MUSIC-based approach is more reliable than the ICA-based approach.

In this paper, Seoulmetro that is the first operation organization which operates a city railroad rolling-stock maintenance RIMS(rolling stock information maintenance system) collected and analyzed a light maintenance data and introduced time a series analysis technique to find the way how to contribute to a use efficiency improvement of a city railroad. The purpose of time a series analysis is to remove a seasonal change including data and to check an irregular fluctuation. First of all, a collection range of the data comes under a light maintenance, however it needs a data of more than 3 years to check the seasonal change. We put a study for an accumulated scope that the data satisfy a period like this and are able to extend a range of the study when time flys forward. The data used for study is filtered using a movement average method after passing proper selection working and is solved with a method which looks for season index. Using the season index that was getten in here, we predict a light working frequency, if it has an irregular change, we will contribute it to a city railroad a use efficiency improvement and establish the cause by carrying out prevent maintenance in advance.

시계열 자료 분석에서 ARCH류와 같은 조건부 이분산성 모형을 가정하고 분석하는 모형들이 많이 쓰이고 있다. 실제 우리나라 금융 시계열 자료들을 분석해 보면 비정상성을 나타내는 경우가 드물지 않게 나타난다. 즉, 단위근 형태의 비정상 패턴(integrated phenomenon)에 가까운 경우가 자주 나타난다. 본 논문에서는 다양한 국내 금융시계열 15개에(주가지수, 선물지수, 환율, 이자율 등) GARCH(1,1) 모형을 적합시켜 분산의 지속성을 확인하고, 각 데이터에 첨도(Kurtosis)와 적합된 IGARCH(1,1) 모형을 제시하고자 한다.

Conditionally heteroscedastic time series models such as GARCH processes have frequently provided useful approximations to the real aspects of financial time series. It is not uncommon that financial time series exhibits near non-stationary, say, integrated phenomenon. For stationary GARCH processes, a shock to the current conditional variance will be exponentially converging to zero and thus asymptotically negligible for the future conditional variance. However, for the case of integrated process, the effect will remain for a long time, i.e., we have a persistent effect of a current shock on the future observations. We are here concerned with providing empirical evidences of persistent GARCH(1,1) for various fifteen domestic financial time series including KOSPI, KOSDAQ and won-dollar exchange rate. To this end, kurtosis and Integrated-GARCH(1,1) fits are reported for each data.

시계열에 대한 색인 및 검색 연구는 하나의 속성으로 구성된 시계열에 대하여 주로 수행되어 왔다. 그러나 음악, 비디오 등의 멀티미디어 데이타베이스는 다중속성 시계열 데이타베이스에서 유사 검색을 다룰 수 있어야 한다. 기존의 다중속성 시계열 데이타베이스에 대한 연구는 두 다중속성 시퀀스간의 유사도로 속성 간의 거리의 누적을 사용하고 있기에, 개별적인 속성 시퀀스에 대한 정보를 상실하게 된다. 본 연구에서는 이러한 문제를 해결하기 위해 속성 시퀀스 측면에서 다중속성 시계열 데이타베이스의 유사검색 기법을 제안한다. 제안된 기법은 검색 공간을 효율적으로 줄일 수 있으며, 착오 누락이 없음을 보장한다. 또한 실험을 통해 제안된 기법의 성능 향상을 확인하였다.

Most of previous work on indexing and searching time series focused on the similarity matching and retrieval of one-attribute time series. However, multimedia databases such as music, video need to handle the similarity search in multi-attribute time series. The limitation of the current similarity models for multi-attribute sequences is that there is no consideration for attributes' sequences. The multi-attribute sequences are composed of several attributes' sequences. Since the users may want to find the similar patterns considering attributes's sequences, it is more appropriate to consider the similarity between two multi-attribute sequences in the viewpoint of attributes' sequences. In this paper, we propose the similarity search method based on attributes's sequences in multi-attribute time series databases. The proposed method can efficiently reduce the search space and guarantees no false dismissals. In addition, we give preliminary experimental results to show the effectiveness of the proposed method.

지하구내에 물이 유입되기 시작하면 미리 설치되어 있는 펌프를 동작시켜서 배수를 해야한다. 이 경우 지하구내로 유입되는 물의 경로로 구분한 내수와 외수의 양을 알 수 있다면 위험도의 평가에 있어서 좋지만 유입량을 정확히 알 수 없으므로 시계열 분석으로 미래의 값을 예측하는 방법을 제시하고자 한다. 시계열 분석으로 예측한 값을 토대로 퍼지 이론을 이용한 지하구내 침수 상황 통제 시스템을 구현하였다.

본 연구에서는 웨이브릿과 신경망을 결합하여 플라즈마 고장을 감시하기 위한 시계열 모델을 개발하였다. 본 기법은 플라즈마 증착장비에 의해 수집된 18 개의 센서정보에 적용하여 평가하였다. 이산치 웨이브릿(Discrete Wavelet Transformation)은 장비에서 수집된 센서정보의 전 처리를 위해 이용되었다. 시계열 모델의 성능은 과거와 미래정보의 함수로 평가하였다. 수집된 18 개의 센서정보에 대한 모델성능 비교를 위해 표준화된 성능평가지표가 적용되었다. 평가결과, 본 기법에 의해 개발된 시계열 모델은 대략 4% 정도의 예측에러를 보였다.

최근 들어 시계열 자료 분석에서 관측된 각 시점에서의 관측치의 분산을 서로 다른 분산(조건부 이분산성)을 따른다고 가정하고, 이를 분석하는 모형(ARCH, GARCH, EGARCH, IGARCH 등)들이 옵션 가격 분석이나 환율 변화 등 경제적 시계열 자료의 예측 모형을 위하여 활발히 연구되고 있다. 본 논문에서는 한국의 KOSPI 데이터 (1999년 1월 4일 $\sim$ 2003년 12월 30일, 총 1227일)를 바탕으로 조건부 우도함수 모수 추정 방법을 이용한 GARCH(1,1), IGARCH(1,1), EGARCH(1,1) 모형에 KOSPI 자료를 적합 시켜 각 모형들의 성능을 비교하여 보았다.

In this paper, we analyse the volatilities in financial data such as stock prices and exchange rates in term of a class of nonlinear time series models. We compare the performance of Generalized Autoregressive Conditional Heteroscadastic(GARCH) , Integrated GARCH(IGARCH), Exponential GARCH(EGARCH) models by KOSPI (Korean stock Prices Index) data. The estimation for the parameters in the models was carried out by the ML methods.