Earticle

The world of linearity, which is regular, predictable and irrelevant to time sequence in most natural phenomenon, is a very small part. In fact, signals generated from natural phenomenon with which we're in contact are showed only slight linearity. Therefore it is very difficult to understand and analyze natural phenomenon with only predictable and regular linear systems. Due to these reasons researches concerning non-linear signals that of analysis were excluded being regarded as noise are being actively carried out. Countless signals generated from nonlinear system have the information about itself, and analyzing those signals and get information from it, that will be able to be used effectively in so may fields. Hence, in this paper we used a higher order spectrum, especially the bispectrum. After we prove the validity applying bispectrum to logistic map, which is typical chaotic signal. Subsequently by showing the result applying for actual signal analysis of EEG according to auditory stimuli, we show that higher order spectra is a very useful parameter in analysis of non-linear signals and the result of EEG analysis according to auditory stimuli.

SX Phe형에 속하는 다주기 변광성 BL Camelopardalis (GD428)에 대하여 전하결합소자(CCD)를 사용한 시계열 차등광전측광을 수행하였다. 관측데이터를 최소자승법과 푸리에 분해(Fourier Decom-position) 방법을 사용하여 주기분석을 시도하였는데 이 변광성의 첫 번째 주기는 0.0391일, 두 번째 주기는 0.0317일 값으로 결정되었고 주기비(PI/PO)는 0.81로 이 변광성은 두 개의 진동양식으로 맥동하는 변광성임을 확인할 수 있었다. 이 주기비 값은 다른 SX Phe형 변광성에 대하여 알려진 값 0.78과 차이가 있고 또한 금속함량비와 주기비 사이의 관계에서도 역시 BL Cam은 다른 SX Phe형 변광성과는 상당한 차이를 보여 이 변광성이 모든 SX Phe형 변광성중에서 매우 극단적인 성질을 갖고 있다는 사실을 확인하였다.

Differential time-series observations of BL Camelopardalis classified as a double mode SX Phoenicis type variable were secured with a charge coupled device. The observed photometric data was reduced using the IRAF Package and the differential magnitudes were obtained through aperture photometry. The periods of BL Cam were analyzed with the Generalized Least-Square Method by Vanicek (1971) and the Fourier Decomposition Method. It was found that the first and second period of BL Cam were 0.0391 day respectively which lead the period ratio of P1/P0=0.81. This period ratio is much different from 0.78 determined by other investigators and also much more larger than that of other double-mode SX Phe type variables. In addition, this period ratio is much different from the value expected from the relation between the metallicity and period ratio. From these results, it can be confirmed that BL Cam is the most extreme case among all double-mode SX Phe type variables.

Wavelet analysis is applying to many fields such as the time-frequency localization of a time series and a time varying data. In this paper, a statistical testing based Wavelet power spectrum analysis for the stationary Nino3 Sea Surface Temperature(SST) data was executed. Specially, the 95% confidence level for SST was effective in searching the periods of El-Nino using various wavelet basis functions.

Travel time forecasting, especially public bus travel time forecasting in urban areas, is a difficult and complex problem which requires a prohibitively large computation time and years of experience. As the network of target area grows with addition of streets and lanes, computational burden of the forecasting systems exponentially increases. Even though the travel time between two neighboring intersections is known a priori, it is still difficult, if not impossible, to compute the travel time between every two intersections. For the reason, previous approaches frequently have oversimplified the transportation network to show feasibilities of the problem solving algorithms. In this paper, forecasting of the travel time between every two intersections is attempted based on travel time data between two neighboring intersections. The time stamps data of public buses which recorded arrival time at predetermined bus stops was extensively collected and forecast. At first, the time stamp data was categorized to eliminate white noise, uncontrollable in forecasting, based on wavelet conversion. Then, the radial basis neural networks was applied to remaining data, which showed relatively accurate results. The success of the attempt was confirmed by the drastically reduced relative error when the nodes between the target intersections increases. In general, as the number of the nodes between target intersections increases, the relative error shows the tendency of sharp increase. The experimental results of the novel approaches, based on wavelet conversion and neural network teaming mechanism, showed the forecasting methodology is very promising.

본 논문은 효과적인 시계열 예측을 위한 새로운 퍼지 학습방법을 제안한다. 기존의 학습방법에서는 입력 데이터를 F(y(t),y(t-1),y(t-2)..)의 형태로 주어 예측을 수행했으나 본 논문에서 제안한 방법에서는 입력 데이터를 F(y(t)-y(t-1),y(t-1)-y(t-2)..)로 설정한다. 이것은 각 입력값의 차이를 새로운 입력으로 사용함으로써 유사한 시계열 분포를 좀더 능동적인 퍼지 규칙으로 만들기 때문에 Non-stationary한 데이터뿐만 아니라 기존의 시계열 데이터 예측방법 보다 나은 결과를 나타낸다. 알고리즘의 수행능력을 살펴보기 위해 Mackey-Glass time series와 Lorenz data를 사용하였다.

본 연구는 거실조명환경을 시계열적으로 분석하기 위하여 1985년, 1991년, 1996년의 조사결과를 고찰한 것이다. 그 결과, 평균조도가 증가되기는 하였으나 아직도 충분한 밝기를 확보하지 못하고 있음이 밝혀졌다. 그리고 조명환경이 향상되었음에도 불구하고 거주자의 의식 및 평가는 그리 높지 않았다.

In this paper, fuzzy logic system equipped with the back-propagation training algorithm as identifiers for nonlinear dynamic systems is described. To improve its performance, Jacob's delta-bar -delta rule is adapted in adjusting stepsize ${\alpha}$, and only y and ${\alpha}$ updating algorithm is suggested. In identifying and predicting the chaotic time series, suggested method is better than Li-Xin Wang's method,[1]

This paper proposes the method to recognize of time series data based on the chaotic feature extraction. Features extract from time series data using the chaotic time series data analysis and the pattern recognition process is using a neural network classifier. In experiment, EEG(electroencephalograph) signals are extracted features by correlation dimension and Lyapunov experiments, and these features are classified by multilayer perceptron neural networks. Proposed chaotic feature extraction enhances recognition results from chaotic time series data.

The purpose of this study is to find the tracking of blood pressure in primary school-age children. A follow-up study was conducted from 1986 to 1990 on 330 first grade children attending primary schools in Kangwha County, Kyungki-Do. Basically we employed a linear regression model with random coefficients to figure out the relation between blood pressure changes and initial blood pressure. We obtained the following results ; 1. The mean blood pressures were increased grade went up in both sexs and were generally higher in female than male except for the systolic blood pressure at first grade. The size of difference was about 0.8 mmHg in mean systolic blood pressure and 1.5 mmHg in mean diastolic blood pressure. 2. The average annual increasing rates of systolic blood pressure were 2.5 mmHg in male and 3.1 mmHg in female respectively. For the diastolic blood pressure IV the average annual increasing rates were observed to be 3.0 mmHg in male and 2.9 mmHg in female respectively. Increasing rate of systolic blood pressure was significantly higher in female than male. 3. The adjusted regression coefficient of systolic blood pressure change on initial value was -0.11 in male and -0.13 in female and that coefficient of diastolic blood pressure change on initial value was -0.01 in male and -0.11 in female. This result shows that children with higher initial blood pressure do not pick up their blood pressure faster than others with lower initial blood pressure. There is no evidence of tracking of blood pressure in children. It is essential to find the earliest age having the tracking of blood pressure and we leave it for the further study.

실시간 스트리밍 환경에서의 시계열 이상치 탐지는 시스템 장애를 조기에 예방하고 운영 리스크를 완화하기 위해 중요하며, 이상치 탐지 연구에는 계절성과 자기상관을 반영할 수 있는 SARIMA 모형이 널리 활용된다. 그러나 기존 SARIMA는 MLE 기반 점추정에 의존하므로 파라미터 불확실성과 예측분포의 변동성을 충분히 반영하지 못해 이상치 판단의 신뢰도와 강건성에 한계가 있다. 베이지안 SARIMA는 이러한 한계를 보완할 수 있으나, 칼만필터의 예측오차 분해 기반 우도를 활용한 전통적 MCMC 추론은 스트리밍 환경에서 데이터 규모가 커질수록 계산 부담이 급격히 증가한다. 본 연구는 이를 해결하기 위해 칼만필터 기반 우도 및 기울기 계산 과정에 Gibbs 샘플링과 확률적 경사 MCMC를 결합한 SGMCMC-KF를 제안하고, 스트리밍 배치 단위로 온라인 사후추론과 예측을 수행하는 이상치 탐지 프레임워크를 제시한다. Apache Kafka 기반 스트리밍 파이프라인에서 잔차 기반 임계값 규칙과 결합하여 다양한 SGMCMC-KF 알고리즘의 탐지 성능을 비교한 결과, SGLD-KF 방법이 MLE 기반 SARIMA 대비 더 안정적인 이상치 탐지 성능을 보였으며, 특히 랑주뱅 노이즈를 통한 확률적 탐색이 스트리밍 환경에서 이상치 민감도 및 강건성 향상에 기여함을 확인하였다.

Detecting anomalies in real-time streaming time series is crucial for early prevention of system failures and mitigation of operational risks. The SARIMA model, which captures seasonality and autocorrelation, is widely used for this task. However, conventional SARIMA relies on MLE based point estimation and cannot adequately account for parameter uncertainty and variability in predictive distributions, limiting the reliability and robustness of anomaly detection. Bayesian SARIMA can address this limitation, but traditional MCMC inference-including Gibbs sampling-based on Kalman filter prediction error decomposition becomes computationally expensive as data scale increases in streaming environments. To address this issue, this study proposes SGMCMC-KF, which combines stochastic gradient MCMC with Kalman filter-based likelihood and gradient computations, and evaluates its performance against Gibbs sampling and MLE-based SARIMA. Experiments in an Apache Kafka-based streaming pipeline show that the SGLD-KF method achieves more stable anomaly detection performance, and that stochastic exploration via Langevin noise improves anomaly sensitivity and robustness in streaming environments.

This paper aims to conduct a future keyword prediction study by applying dynamic topic modeling and the VAR model through time series analysis using artificial intelligence-related patent data from 2012 to 2022. Patent data was collected through the KIPRIS platform, and major keywords such as ‘information’, ‘data’, and ‘user’ were constantly dealt with in the annual topic trend graph derived through dynamic topic modeling, and the importance of topics such as ‘image’ and ‘video’ has been increasing in recent years. These results suggest that research trends in various application fields are changing with the development of artificial intelligence technology. Next, 10, 20, and 30 top keywords were extracted respectively from the yearly patent data collected for future keyword prediction using BoW, TF-IDF, and Word2Vec, which are three embedding methods. It was confirmed that ‘information’, ‘data’, ‘user’, ‘device’, and ‘system’ were the same important keywords in the annual top keyword analysis by year. The ranking of different patterns was subsequently shown for each embedding method. After that, an experiment was conducted to compare the keywords predicted through the VAR model with the actual 2023 keywords, and it was confirmed that the Word2Vec embedding method showed the highest prediction performance, and the overall performance improved as the number of keywords increased. This study provided important basic data for predicting technology trends and future trends through artificial intelligence patent data, and it is expected that it can be used as a basis for strategic use directions and investment decisions for various stakeholders such as companies, research institutes, and the government.

수출입 화물수송은 교역의 크기를 나타내는 대표적 지표이며 그 중 해운 화물수송은 전체 수출입 화물수송 중 가장 큰 비중을차지한다. 본 연구는 한국의 해운 화물수송량의 동태적 추이를 장기기억의 반모수 검정 기법을 통해 살펴보고자 한다. 장기기억검정은 기존 접근에 비해 보다 정확하게 비정상 시계열 여부를 판별할 수 있으며, 본 연구는 이러한 장기기억의 반모수 검정을통해 한국 해운 화물수송량의 동태적 변화를 고찰하고자 한다. 또한, 보다 정확한 장기기억 검정을 위해 본 연구는 시계열 자료에구조적 변화가 존재할 가능성을 검정 기법에 반영한다. 본 연구는 해운 화물수송을 수출과 수입으로 분류하여 각각에 대해 장기기억 검정을 적용한다. 본 연구의 실증분석 결과에 따르면 총 수출 화물수송량과 총 수입 화물수송량의 구조적 변화 시점은 상이한반면, 장기기억 계수값은 유사하다. 본 연구에 제시된 장기기억 검정 결과에 따르면 총 수출 화물수송량과 총 수입 화물수송량 모두 전통적인 비정상 시계열에 비해 충격 지속성이 낮으며 충격 지속성은 수출과 수입에 있어서 유사한 것으로 나타났다. 한편, 자동차 부문의 수출 화물수송량은 수입 화물수송량에 비해 충격이 상대적으로 더 지속적인 것으로 나타났으며 전기기기 부문의 경우 수입 화물수송량의 충격이 수출 화물수송량의 충격에 비해 더 지속적인 것으로 나타났다. 본 연구의 실증분석 결과에 따르면전체 수출 품목의 화물수송량과 전체 수입 품목의 화물수송량의 동태적 변화는 구조적 변화를 통제했을 때 유사한 반면 자동차, 전기기기 등 주요 품목의 수출입 화물수송량은 구조적 변화를 감안한 후에도 수출과 수입 사이에 동태적 변화가 상이한 것으로나타났다.

Freight transport for exports and imports is a crucial indicator of trade volume. In particular, maritime freight transport accounts for the largest proportion of total export and import freight transport for Korea. This study examines the dynamic trends in Korea’s maritime freight transport volume using semiparametric tests of long memory. Long memory tests can more accurately determine whether a time series is non-stationary compared to traditional approaches. This study uses these semiparametric long memory tests to investigate the dynamic changes in Korea’s maritime freight transport volume. Additionally, to enhance the accuracy of the long memory tests, this study incorporates the possibility of structural changes in the time series data into the long memory test. This study classifies maritime freight transport into exports and imports and applies long memory tests to each category. According to the results, the structural change points for total export freight transport volume and total import freight transport volume differ, whereas the long memory coefficient values are similar. The long memory test results of this study indicate that the persistence of shocks for both total export and total import freight transport volumes is lower than unit root non-stationary time series, while the degree of persistence is similar for both exports and imports. On the other hand, the export freight transport volume in the automobile sector shows relatively more persistent shocks compared to import freight transport volume. In the electric machinery sector, import freight transport volume shocks are more persistent than export freight transport volume shocks. The results of this study show that the dynamic changes in the freight transport volume of total export items and total import items are similar when controlling for structural changes. In contrast, the dynamic changes in the freight transport volumes of major items such as automobiles and electric machinery differ between exports and imports even after accounting for structural changes.

회계정보의 시계열적 비교가능성(이하 ‘회계일관성’)은 회계실무와 이론 모두에서 그 중요성이 강조되는 회계정보의 질적 특성이다. 본 연구에서는 ‘특정기업의 상이한 기간에 일어난 동일한 경제적인 사건이 동일한 회계수치로 표출되는 정도’로 회계일관성을 정의한 후 (1) 특정 기업의 회계일관성이 기업 간 비교가능성과 관련이 있는지, (2) 회계일관성이 감사인의 특성과 관련이 있는지를 조사하였다. 국내 상장기업의 2004-2017년 자료를 사용하여 조사한 연구결과는 다음과 같다. 첫째, 재무제표 비교가능성과 회계일관성은 양(+)의 관련성을 가졌다. 이는 다른 기업과 비교가능한 재무제표를 작성하는 기업은 여러 기간에 걸쳐 일관성 있는 회계수치를 보고하는 경향이 있음을 의미한다. 둘째, 회계일관성 측정기간에 걸쳐 동일한 감사인으로부터 감사를 받은 피감사기업은 그렇지 않은 피감사기업보다 회계일관성이 높은 것으로 나타났다. 동일감사인이 계속감사를 하면 이 감사인은 대체로 일관성 있는 기준이나 해석을 적용할 가능성이 높다는 것을 감안하면 이 결과는 예상과 일치한다. 셋째, 계속감사를 한 감사인이 고품질감사인(Big4 또는 산업전문감사인)인지 여부는 회계일관성에 증분적인 영향을 미치지 않는 것으로 나타났다. 본 연구의 발견은 (감사인의 품질이 아닌) 동일감사인으로부터 계속감사를 받았는지 여부가 회계일관성을 결정하는 핵심요소임을 시사한다.

This study investigates (1)whether time-series accounting consistency of a firm is related to cross-sectional comparability, and (2)the relationship between accounting consistency and the characteristics of auditors. To do so, we redefine the accounting consistency as "the extent to which identical economic events occurring in different periods for a particular company are represented by similar accounting figures". The findings using the data of Korean publicly-listed companies from 2004 to 2017 are as follows. Firstly, accounting consistency and cross-sectional comparability have a positive relationship. This suggests that companies that consistently report financial figures across multiple periods produce financial statements that are comparable to those of peer companies. Secondly, companies audited by the same auditor tend to exhibit higher levels of accounting consistency compared to those companies audited by different auditors. Lastly, whether the auditor is a high-quality auditor (e.g., Big4 or industry specialist auditor) has no systematic impact on accounting consistency. Overall, our findings suggest that the key determinants of time-series accounting consistency is whether the same auditor have conducted continuous audits, rather than the quality of the auditor.

본 연구는 ESG와 음식료품 산업 시계열 지수 분석을 통하여 상호 미친 영향을 실증적으로 파악하고자 한다. 본 연구에서 사용한 자료는 2010년 1월 9일부터 2023년 4월 8일까지의 692개의 ESG 지수와 음식료품 지수의 주간 시계열 자료이다. 이를 이용하여 “예수께서 떡 다섯 개와 물고기 두 마리를 가지사 하늘을 우러러 축사하시고 떡을 떼어 제자들에게 주어 사람들에게 나누어 주게 하시고 또 물고기 두 마리도 모든 사람에게 나누시매 다 배불리 먹고 남은 떡 조각과 물고기를 열두 바구니에 차게 거두었으며 떡을 먹은 남자는 오천 명이었더라”의 마가복음 6장 말씀에서 오병이어의 표적뿐만 아니라 남은 음식물을 모두 처리하여 환경오염을 막았다. 이에 따라 현재의 음식료품과 ESG와의 관계를 실증적으로 살펴보고자 한다. ESG와 음식료품 산업에 상호 미친 영향을 살펴보고 ESG 지수와 음식료품 지수의 주간 시계열 자료 간의 영향력 정도를 계량경제분석모형을 이용하여 분석하고자 한다. ESG와 음식료품 산업에 관한 실증적인 분석의 주요결과는 다음과 같다. 첫째, ESG 지수와 음식료품 지수의 주간 시계열 자료의 안정성검정 결과 수준변수는 불안정적인 것으로 나타났고 차분변수는 안정성 검정결과는 안정적임을 알 수 있었다. 둘째, ESG 지수와 음식료품 지수의 주간 시계열 자료의 공적분 관계가 수준변수 간에는 존재하지 않으나 차분변수에서는 공적분 관계가 존재하는 것으로 나타났다. 셋째, ESG 지수와 음식료품 지수의 주간 시계열 자료의 상관관계는 0.308267로 양(+)의 관계로 나타났다. 마지막으로 음식료품 지수는 ESG 지수에 영향을 미치는 원인변수로 나타나 Granger 인과관계가 있는 것으로 분석되었다. ESG와 음식료품 산업에 상호 미친 영향을 실증적으로 분석한 연구의 한계점으로는 정확하고 다양한 ESG와 음식료품의 공개된 자료가 존재하지 않는다는 것이다.

This study aims to empirically analyze the mutual impact of ESG and food and beverage industry. The data used in this study are 692 ESG index and F&B index weekly time series data from January 9, 2010 to April 8, 2023, “Jesus took five loaves and two fish, looked up to heaven, and gave a congratulatory speech. Then he broke the loaves and gave them to the disciples to distribute to the people. He also gave two fish to everyone, and they all ate and were satisfied. Then they took up twelve baskets full of the leftover pieces of bread and fish, and the men who ate the loaves were five thousand.” According to the treatment of food in Chapter 6, we will empirically examine the relationship between food and beverage and ESG. By examining the mutual impact of ESG and food and beverage industry, we will analyze the degree of influence between the weekly time series data of food and beverage index and ESG index using an econometric analysis model. The main results of the empirical analysis on ESG and food and beverage industry are as follows. First, it was found that there was no cointegration relationship between ESG index and the weekly time series data of F&B index, but there was a cointegration relationship between the level variables, which are the first log difference data. Second, the correlation between the weekly time series data of ESG index and F&B index was 0.308267, showing a positive (+) relationship. Finally, food and beverage index was found to be a causal variable affecting ESG index, and it was analyzed that there was a Granger causal relationship. A limitation of studies that empirically analyze the mutual impact of ESG on food and beverage industry is that there is no accurate ESG index.