• 응용통계연구
• /
• 제36권5호
• /
• pp.457-471
• /
• 2023

월별 시계열은 일별 시계열의 월별 합이지만, 일별 시계열을 대체로 관측할 수 없어서 요일구성변동, 명절·공휴일변동 등 달력변동을 가상적으로 가정한 가변수를 포함한 RegARIMIA 모형을 이용하여 추정하고 있다. 일별 시계열을 관측할 수 있다면 요일구성변동, 명절·공휴일변동 등 달력변동을 일별 시계열을 바탕으로 추정할 수 있고 이를 이용하여 월별 시계열의 계절조정을 개선할 수 있다. 이 논문에서는 일별 시계열의 달력변동 추정을 이용하여 월별 시계열의 계절조정을 개선하는 방법을 제안하고, 이 방법을 적용하여 3개의 월별 시계열을 계절조정하고 기존의 X-13ARIMA-SEATS를 이용한 계절조정과 비교하였다.

• 한국수자원학회논문집
• /
• 제39권3호
• /
• pp.253-259
• /
• 2006

기존의 거의 모든 수문학적 연구에 있어서, 시스템의 특성을 파악한 뒤 예측을 실시하는 표준접근법이 채택되어왔다. 그러나 최근 들어 시스템의 특성분석에 앞서 예측을 실시하고, 상태공간 매개변수가 시스템의 특성분석단계가 아닌 예측단계에서 평가되는 가역접근법이 제안되었다. 본 연구에서는 최근에 제안된 가역접근법과 기존에 널리 적용되어온 표준접근법을 이론적 카오스 시계열과 Idaho주 Bear강의 일유출량 자료에 적용함으로써, 가역접근법의 적용성을 검토하고 카오스 시계열의 특성을 알아보았으며, 카오스이론이 적용된 비선형 예측기법으로는 부분근사화 기법을 이용하였다. 카오스 특성 분석을 통해, 이론적 카오스 시계열과 Idaho주 Bear강의 일유출량 시계열 자료 모두에서 카오스 특성이 나타남을 알 수 있었다. 200일에 대한 1, 3, 5일 예측 결과, 가역접근법이 표준접근법에 비해 우수함을 알 수 있었다.

• 한국환경과학회지
• /
• 제18권4호
• /
• pp.369-374
• /
• 2009

To examine the trend on the flowering time in some weather flora including Prunus serrulata var. spontanea, Cosmos bipinnatus, and Robinia pseudo-acacia in Busan, the changes in time series and rate of flowering time of plants were analyzed using the method of time series analysis. According to the correlation between the flowering time and the temperature, changing pattern of flowering time was very similar to the pattern of the temperature, and change rate was gradually risen up as time goes on. Especially, the change rate of flowering time in C. bipinnatus was 0.487 day/year and showed the highest value. In flowering date in 2007, the difference was one day between measurement value and prediction value in C. bipinnatus and R. pseudo-acacia, whereas the difference was 8 days in P. mume showing great difference compared to other plants. Flowering time was highly related with temperature of February and March in the weather flora except for P. mume, R. pseudo-acacia and C. bipinnatus. In most plants, flowering time was highly related with a daily average temperature. However, the correlation between flowering time and a daily minimum temperature was the highest in Rhododendron mucronulatum and P. persica, otherwise the correlation between flowering time and a daily maximum temperature was the highest in Pyrus sp.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1995년도 하계학술대회 논문집 B
• /
• pp.578-580
• /
• 1995

In this paper, a method for the daily maximum load forecasting which uses a chaotic time series in power system and artificial neural network. We find the characteristics of chaos in power load curve and then determine a optimal embedding dimension and delay time, For the load forecast of one day ahead daily maximum power, we use the time series load data obtained in previous year. By using of embedding dimension and delay time, we construct a strange attractor in pseudo phase plane and the artificial neural network model trained with the attractor font mentioned above. The one day ahead forecast errors are about 1.4% of absolute percentage average error.

• Journal of the Korean Data and Information Science Society
• /
• 제16권3호
• /
• pp.695-703
• /
• 2005

This paper extends the maximum likelihood seasonal cointegration procedure developed by Johansen and Schaumburg (1999) for daily time series. The finite sample distribution of the associated rank test for dally data is also presented.

• 한국농공학회지
• /
• 제31권4호
• /
• pp.72-80
• /
• 1989

Various analyses were made to investigate the stochastic structure of the daily rainfall in Korea. Records of daily rainfall amounts from 1951 to 1984 at Chinju Metesrological Station were used for this study. Obtained results are as follows : 1. Time series of the daily rainfall at Chinju were positively, serially correlated for the lag as large as one day. 2. Rainfall events, defined as a sequence of consecutive wet days separated by one or more dry days, showed a seasonal variation in the occurrence frequency. 3. The marginal distribution of event characteristics of each month showed significant dif- ferences each other. Events occurred in summer had longer duration and higher magnitude with higher intensity than those of events occurred in winter. 4. There were significant positive correlations among four event characteristics ; dura- tion, magnitude, average intensity, and maximum intensity. 5. Correlations among the daily rainfall amounts within an event were not significant in general. 6. There were no consistant significancy in identity or difference between the distribu- tions of daily rainfall amounts for different days within events. 7. Above mentioned characteristics of daily rainfall time series must be considered in building a stochastic model of daily rainfall.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2010년도 학술발표회
• /
• pp.1439-1442
• /
• 2010

Time-lag recurrent neural networks model (Time-Lag RNNM) is used to estimate daily pan evaporation (PE) using limited climatic variables such as max temperature ($T_{max}$), min temperature ($T_{min}$), mean wind speed ($W_{mean}$) and mean relative humidity ($RH_{mean}$). And, for the performances of Time-Lag RNNM, it is composed of training and test performances, respectively. The training and test performances are carried out using daily time series data, respectively. From this research, we evaluate the impact of Time-Lag RNNM for the modeling of the nonlinear time series data. We should, thus, construct the credible data of the daily PE using Time-Lag RNNM, and can suggest the methodology for the irrigation and drainage networks system. Furthermore, this research represents that the strong nonlinear relationship such as pan evaporation modeling can be generalized using Time-Lag RNNM.

• 한국습지학회지
• /
• 제21권3호
• /
• pp.236-243
• /
• 2019

현재까지 많은 수문 시계열은 전통적인 선형 모형을 이용하여 분석되고 예측되어 왔다. 하지만, 자연현상과 수문시계열의 패턴 및 변동과 관련하여 비선형적 구조의 증거가 발견되고 있다. 따라서 시계열 분석 및 예측을 위한 기존의 선형 모형은 비선형적 특성에 적합하지 않을 수 있다. 본 연구에서는 미국 플로리다 코코아 지역 인근에 있는 St.Johns 강의 일유량 자료에 대한 카오스 분석을 수행하였고, 그 결과 낮은 차원의 비선형 동역학적 특성을 가진 흥미로운 결과가 나타났지만 한국의 소양강댐 일유량 자료는 확률적 특성을 보여주었다. 카오스 특성을 토대로한 DVS(결정론적 vs 추계학적) 알고리즘을 이용해 두 시계열 시스템의 특성을 파악하였고 단기 예측을 수행하였다. 또한 본 연구에서는 일 유량 시계열 예측을 위해 인공신경망 방법을 사용하였고, DVS 알고리즘에 의한 예측을 비교 분석하였다. 분석 결과, 카오스 특성을 갖는 시계열 자료가 보다 정확한 예측성을 보였다.

• 상하수도학회지
• /
• 제18권6호
• /
• pp.758-769
• /
• 2004

The trends and seasonalities of most time series have a large variability. The result of the Singular Spectrum Analysis(SSA) processing is a decomposition of the time series into several components, which can often be identified as trends, seasonalities and other oscillatory series, or noise components. Generally, forecasting by the SSA method should be applied to time series governed (may be approximately) by linear recurrent formulae(LRF). This study examined forecasting ability of SSA-LRF model. These methods are applied to daily water demand data. These models indicate that most cases have good ability of forecasting to some extent by considering statistical and visual assessment, in particular forecasting validity shows good results during 15 days.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1996년도 하계학술대회 논문집 B
• /
• pp.835-837
• /
• 1996

In this paper, a method for the daily maximum load forecasting which uses a chaotic time series in power system and artificial neural network(Back-propagation) is proposed. We find the characteristics of chaos in power load curve and then determine a optimal embedding dimension and delay time. For the load forecast of one day ahead daily maximum power, we use the time series load data obtained in previous year. By using of embedding dimension and delay time, we construct a strange attractor in pseudo phase plane and the artificial neural network model trained with the attractor mentioned above. The one day ahead forecast errors are about 1.4% for absolute percentage average error.