• 산경연구논집
• /
• 제8권4호
• /
• pp.15-21
• /
• 2017

Purpose - In this work, we examined the causal relationship between credit loans from households (CLH), loan collateralized with housing (LCH) and an interest of certificate of deposit (ICD) among others in South Korea. Furthermore, the optimal forecasts on the underlying model will be obtained and have the potential for applications in the economic field. Research design, data, and methodology - A total of 31 realizations sampled from the 4th quarter in 2008 to the 4th quarter in 2016 was chosen for this research. To achieve the purpose of this study, a regression model with correlated errors was exploited. Furthermore, goodness-of-fit measures was used as tools of optimal model-construction. Results - We found that by applying the regression model with errors component ARMA(1,5) to CLH, the steep and lasting rise can be expected over the next year, with moderate increase of LCH and ICD. Conclusions - Based on 2017-2018 forecasts for CLH, the precipitous and lasting increase can be expected over the next two years, with gradual rise of two major explanatory variables. By affording the assumption that the feedback among variables can exist, we can, in the future, consider more generalized models such as vector autoregressive model and structural equation model, to name a few.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2015년도 학술발표회
• /
• pp.225-225
• /
• 2015

Drought events usually evolve slowly in time and their impacts generally span a long period of time. This indicates that the sequence of drought is not completely random. The Hidden Markov Model (HMM) is a probabilistic model used to represent dependences between invisible hidden states which finally result in observations. Drought characteristics are dependent on the underlying generating mechanism, which can be well modelled by the HMM. This study employed a HMM with Gaussian emissions to fit the Standardized Precipitation Index (SPI) series and make multi-step prediction to check the drought characteristics in the future. To estimate the parameters of the HMM, we employed a Bayesian model computed via Markov Chain Monte Carlo (MCMC). Since the true number of hidden states is unknown, we fit the model with varying number of hidden states and used reversible jump to allow for transdimensional moves between models with different numbers of states. We applied the HMM to several stations SPI data in South Korea. The monthly SPI data from January 1973 to December 2012 was divided into two parts, the first 30-year SPI data (January 1973 to December 2002) was used for model calibration and the last 10-year SPI data (January 2003 to December 2012) for model validation. All the SPI data was preprocessed through the wavelet denoising and applied as the visible output in the HMM. Different lead time (T= 1, 3, 6, 12 months) forecasting performances were compared with conventional forecasting techniques (e.g., ANN and ARMA). Based on statistical evaluation performance, the HMM exhibited significant preferable results compared to conventional models with much larger forecasting skill score (about 0.3-0.6) and lower Root Mean Square Error (RMSE) values (about 0.5-0.9).

• 지능정보연구
• /
• 제5권1호
• /
• pp.103-123
• /
• 1999

Input filtering as a preprocessing method is so much crucial to get good performance in time series forecasting. There are a few preprocessing methods (i.e. ARMA outputs as time domain filters, and Fourier transform or wavelet transform as time-frequency domain filters) for handling time series. Specially, the time-frequency domain filters describe the fractal structure of financial markets better than the time domain filters due to theoretically additional frequency information. Therefore, we, first of all, try to describe and analyze specially some issues on the effectiveness of different filtering methods from viewpoint of the performance of a neural network based forecasting. And then we discuss about neural network model architecture issues, for example, what type of neural network learning architecture is selected for our time series forecasting, and what input size should be applied to a model. In this study an input selection problem is limited to a size selection of the lagged input variables. To solve this problem, we simulate on analyzing and comparing a few neural networks having different model architecture and also use an embedding dimension measure as chaotic time series analysis or nonlinear dynamic analysis to reduce the dimensionality (i.e. the size of time delayed input variables) of the models. Throughout our study, experiments for integration methods of joint time-frequency analysis and neural network techniques are applied to a case study of daily Korean won / U. S dollar exchange returns and finally we suggest an integration framework for future research from our experimental results.

• 물과 미래
• /
• 제29권2호
• /
• pp.179-190
• /
• 1996

한강하류부 수질의 통계학적 해석을 통하여 수질 시계열자료의 기본 통계특성치, 지점별 및 계절별 변동성을 검토하였으며, 유량과 수질인자간의 상관성 분석을 실시하였다. 본류의 주요 6개 지점 및 3개 지류에 대한 통계특성치와 적정분포형을 산정하여 제시하였으며, 시간의존성 및 계절성을 검토하여 제시하였다. 또한, 수질 항목간의 상관성 검토를 통하여 상관성이 높은 수질, 항목간, 그리고 지점간의 상관식을 제시하였다. 추계학적 모의모형의 적용가능성을 확인하였으며, DO 항목은 전 지점간에 높은 상관성을 가지고 있었다. 유량과의 상관관계 검토에 있어서 DO, SS 항목은 유량보다는 수온에 민감하였으며, BOD, COD 항목은 유량이 적은 갈수기에는 유량에 민감한 것으로 나타났다. 수온에 밀접한 영향을 받는 DO 항목외에도 BOD, COD 항목은 계절적인 주기성을 가지고 있었으며, 상호상관 분석결과 DO, BOD, COD 항목 외의 수질 항목들에서도 각 수질 항목들에 내재된 주기성을 찾아볼 수 있었다.

• 한국농림기상학회지
• /
• 제9권2호
• /
• pp.88-99
• /
• 2007

토양수분은 토양으로의 침투나 지표유출 기작에 직접적인 영향을 주며 간접적으로 유역 단위의 수문학적, 수리화학적, 기상학적, 생태학적 반응에 중요한 역할을 한다. 본 연구에서는 광릉 슈퍼사이트 내 원두부 소유역을 대상으로 사면에서의 토양수분 전이과정이 토양수분의 추계학적 모형구조와 갖는 연계성을 규명하기 위해서 일련의 유도과정이 수행되었다. 유도된 단변량 추계학적 모형의 구조에 근거하여, 관측된 토양수분의 시계열을 모의하였다. 자료전처리, 모형구조의 규명, 후보 모형군의 구성, 모수추정, 검정 등의 과정을 통해서 도출된 모형들의 공간적인 분포는 대상사면의 지형학적인 특성들이 반영된 것으로 판단된다. 다방향 알고리즘에 의한 기여면적이나 습윤 지수와 함께 대상지점의 국부경사도가 중요변수로 도출되었다. 본 연구 결과는 광릉 슈퍼사이트와 같은 복잡 경관에서 토양수분의 공간분포를 결정짓는 중요한 요인들을 이해하고 이를 통해 현실성있는 토양수분 분포 지도를 작성하는데 기여 하게 될 것이다.