• Proceedings of the KIEE Conference
• /
• 2005.07a
• /
• pp.819-821
• /
• 2005

Since the System Marginal Price (SMP) is a vital factor to the market entities who intend to maximize the their profit, the short-term marginal price forecasting should be performed correctly. In a electricity market, the short-term trading between the market entities can be generally affected a short-term market price. Therefore, the exact forecasting of SMP can influence on the profit of market participants. This paper presents a methodology of day-ahead SMP foretasting using Autoregressive Integrated Moving Average (ARIMA). To show the efficiency and effectiveness of the proposed method, the numerical studies have been performed using historical data of SMP in 2004.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.7
• /
• pp.11-18
• /
• 2023

In this paper, we evaluate deep learning time series forecasting models. Recent studies show that those models perform better than the traditional prediction model such as ARIMA. Among them, recurrent neural networks to store previous information in the hidden layer are one of the prediction models. In order to solve the gradient vanishing problem in the network, LSTM is used with small memory inside the recurrent neural network along with BI-LSTM in which the hidden layer is added in the reverse direction of the data flow. In this paper, we compared the performance of Informer by comparing with other models (LSTM, BI-LSTM, and Transformer) for real Nitrogen dioxide (NO2) data. In order to evaluate the accuracy of each method, mean square root error and mean absolute error between the real value and the predicted value were obtained. Consequently, Informer has improved prediction accuracy compared with other methods.

• Journal of Digital Convergence
• /
• v.16 no.8
• /
• pp.169-179
• /
• 2018

Recently, web search query information has been applied in advanced predictive model research. Google dominates the global web search market in the Korean market; however, Naver possesses a dominant market share. Based on this characteristic, this study intends to compare the utility of the Korean web search query information of Google and Naver using predictive models. Therefore, this study develops three time-series predictive models to estimate the youth unemployment rate in Korea using the ARIMA model. Model 1 only used the youth unemployment rate in Korea, whereas Models 2 and 3 added the Korean web search query information of Naver and Google, respectively, to Model 1. Compared to the predictability of the models during the training period, Models 2 and 3 showed better fit compared with Model 1. Models 2 and 3 correlated different query information. During predictive periods 1 (continuous with the training period) and 2 (discontinuous with the training period), Model 3 showed the best performance. During predictive period 2, only Model 3 exhibited a significant prediction result. This comparative study contributes to a general understanding of the usefulness of Korean web query information using the Naver and Google search engines.

• Journal of Internet of Things and Convergence
• /
• v.6 no.4
• /
• pp.7-14
• /
• 2020

This study analyzed the effect of the number of deaths of circulatory system diseases according to 12-day short-term exposure of carbon monoxide from January 2010 to December 2018, and predicted the future treatment cost of circulatory system diseases according to increased carbon monoxide concentration. Data were extracted from Air Korea of Korea Environment Corporation and Korea Statistical Office, and analyzed using Poisson regression analysis and ARIMA intervention model. For statistical processing, SPSS Ver. 21.0 program was used. The results of the study are as follows. First, as a result of analyzing the relationship between the impact of short-term carbon monoxide exposure on death of circulatory system diseases from the day to the previous 11 days, it was found that the previous 11 days had the highest impact. Second, with the increase in carbon monoxide concentration, the future circulatory system disease treatment cost was estimated at 10,123 billion won in 2019, higher than the observed value of 9,443 billion won at the end of December 2018. In addition, when summarized by month, it can be seen that the cost of treatment for circulatory diseases increases from January to December, reflecting seasonal fluctuations. Through such research, the future for a healthy life for all citizens can be realized by distributing various devices and equipment utilizing IoT to preemptively respond to the increase in air pollutants such as carbon monoxide.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2018.05a
• /
• pp.11-11
• /
• 2018

본 연구에서는 1시간부터 1년 단위의 강우 특성들을 잘 모의하는 혼합 추계 강우 생성 모형을 개발하였다. 본 모형의 가상 강우 생성 과정은 4단계로 이루어진다. 첫 단계에서 Seasonal ARIMA 모형을 통하여 시계열 특성을 반영한 월 강우를 생성한다. 두 번째 단계는 생성된 월 강우에 해당하는 일 단위 이하의 강우 통계치 세트를 생성하는 것이며, 통계치간 상관관계를 통해 평균, 표준편차, 자기상관 계수, 무강우 확률을 생성한다. 생성된 통계치 세트는 세 번째 단계에서 Modified Bartlett-Lewis Rectangular Pulse (MBLRP) 모형의 6개의 매개변수를 보정하는데 사용되며, 마지막으로 MBLRP 매개변수 세트를 통해 가상 강우 시계열을 생성한다. 위 모형을 통해 미국 동부 지역 29개 강우 관측소에 대하여 200년 길이의 가상 강우를 생성하였으며, 그 결과 시 단위부터 연 단위까지 강우의 1차, 2차 통계치 및 무강우 확률을 성공적으로 재현하였다. 또한 기존 MBLRP 모형에 비하여 극한 강우 사상을 재현하는 능력이 향상되었다. 빈도분석 결과를 통하여 MBLRP 모형이 재현기간에 따라 10%에서부터 40%까지 극한 사상을 과소 추정한 반면, 본 모형에서는 20% 이내의 값을 나타내었다.

• Journal of the Korean Data and Information Science Society
• /
• v.24 no.6
• /
• pp.1103-1112
• /
• 2013

Exploratory data analysis focuses mostly on data exploration instead of model fitting. We can use the recurrence plot as a graphical exploratory data analysis tool. With the recurrence plot, we can obtain the structural pattern of the time series and recognize the structural change points in time series at a glance.

• Proceedings of the Korea Water Resources Association Conference
• /
• 1984.07a
• /
• pp.89-98
• /
• 1984

우리나라의 월강우량 기록은 풍부하나 월유출량 기록은 희박하여, 월유출량 시계열의 모형식을 개발하고저 하여 월강우량 기록만으로 하천유량의 정확한 파악을 할 수 있도록 한다. 이 연구는 월강우와 유출량의 시계열에 의한 추계학적 이론에 의거한 복스와 젠킨스의 대체함수(Transfer function model)와 아리마(ARIMA)의 잔차모양을 합한 형이다. 이 선형 추계학적 차분 시계열식 모형은 공본산(coveriance) 을 갖는다는 가정에서 강우량과 유출량의 변화에 따라서 식의 구조가 유도되며 정확하게 잘 적용이 된다. 본 식의 최적모형은 일반식으로 아래와 같이 얻어진다. $ Y$:월유출량, X$:월강우량, C$:유출물, $: 대체변수, a$:백색잡음(white noise), $\theta$(B) 및 (B):MA(Moving average)와 AR(autoregressive)조작, 이번 연구 결과 섬진강 하천의 대체조작(Transfer operator)은 잔차승(Sum of residual) R$0.9로 높은 정도의 수치를 나타내는 것으로 보인다.

• The Korean Journal of Applied Statistics
• /
• v.34 no.5
• /
• pp.723-734
• /
• 2021

Wind energy is one of the rapidly developing renewable energies which is being developed and invested in response to climate change. As renewable energy policies and power plant installations are promoted, the supply of wind power in Korea is gradually expanding and attempts to accurately predict demand are expanding. In this paper, the ARIMA and ARIMAX models which are Time series techniques and the SVR, Random Forest and XGBoost models which are machine learning models were compared and analyzed to predict wind power generation in the Jeonnam and Gyeongbuk regions. Mean absolute error (MAE) and mean absolute percentage error (MAPE) were used as indicators to compare the predicted results of the model. After subtracting the hourly raw data from January 1, 2018 to October 24, 2020, the model was trained to predict wind power generation for 168 hours from October 25, 2020 to October 31, 2020. As a result of comparing the predictive power of the models, the Random Forest and XGBoost models showed the best performance in the order of Jeonnam and Gyeongbuk. In future research, we will try not only machine learning models but also forecasting wind power generation based on data mining techniques that have been actively researched recently.

• Communications for Statistical Applications and Methods
• /
• v.19 no.3
• /
• pp.507-516
• /
• 2012

We study time series models for seasonal time series data with a covariance structure that depends on time and the periodic autocorrelation at various lags $k$. In this paper, we introduce an ARMA model with periodically varying coefficients(PARMA) and analyze Arosa ozone data with a periodic correlation in the practical case study. Finally, we use a PARMA model and a seasonal ARIMA model for data analysis and show the performance of a PARMA model with a comparison to the SARIMA model.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.6
• /
• pp.991-998
• /
• 2009

This paper deals with directivity forecasting of time series which is useful for futures trading in stock market. Directivity forecasting of time series is to forecast whether a given time series will rise or fall at next observation time point. For directional forecasting, we consider time regression model and ARIMA model. In particular, we study two statistics, intra-model and extra-model deviation and then show usefulness of intra-model deviation.