• The Korean Journal of Applied Statistics
• /
• v.31 no.6
• /
• pp.785-799
• /
• 2018

With the development of renewable energy sector, the importance of solar energy is continuously increasing. Solar radiation forecasting is essential to accurately solar power generation forecasting. In this paper, we used time series models (ARIMA, ARIMAX, seasonal ARIMA, seasonal ARIMAX, ARIMA GARCH, ARIMAX-GARCH, seasonal ARIMA-GARCH, seasonal ARIMAX-GARCH). We compared the performance of the models using mean absolute error and root mean square error. According to the performance of the models without exogenous variables, the Seasonal ARIMA-GARCH model showed better performance model considering the problem of heteroscedasticity. However, when the exogenous variables were considered, the ARIMAX model showed the best forecasting accuracy.

• The Journal of Information Systems
• /
• v.29 no.1
• /
• pp.241-265
• /
• 2020

Purpose Various machine learning techniques are used to implement for predicting corporate credit. However, previous research doesn't utilize time series input features and has a limited prediction timing. Furthermore, in the case of corporate bond credit rating forecast, corporate sample is limited because only large companies are selected for corporate bond credit rating. To address limitations of prior research, this study attempts to implement a predictive model with more sample companies, which can adjust the forecasting point at the present time by using the credit score information and corporate information in time series. Design/methodology/approach To implement this forecasting model, this study uses the sample of 2,191 companies with KIS credit scores for 18 years from 2000 to 2017. For improving the performance of the predictive model, various financial and non-financial features are applied as input variables in a time series through a sliding window technique. In addition, this research also tests various machine learning techniques that were traditionally used to increase the validity of analysis results, and the deep learning technique that is being actively researched of late. Findings RNN-based stateful LSTM model shows good performance in credit rating prediction. By extending the forecasting time point, we find how the performance of the predictive model changes over time and evaluate the feature groups in the short and long terms. In comparison with other studies, the results of 5 classification prediction through label reclassification show good performance relatively. In addition, about 90% accuracy is found in the bad credit forecasts.

• Journal of IKEEE
• /
• v.23 no.1
• /
• pp.127-133
• /
• 2019

Electric power demand forecasting is one of the important research areas for future smart grid introduction. However, It is difficult to predict because it is affected by many external factors. Traditional methods of forecasting power demand have been limited in making accurate prediction because they use raw power data. In this paper, a probability-based CRBM is proposed to solve the problem of electric power demand prediction using raw power data. The stochastic model is suitable to capture the probabilistic characteristics of electric power data. In order to compare the mid-term power demand forecasting performance of the proposed model, we compared the performance with Recurrent Neural Network(RNN). Performance comparison using electric power data provided by the University of Massachusetts showed that the proposed algorithm results in better performance in mid-term energy demand forecasting.

• Architectural research
• /
• v.8 no.1
• /
• pp.77-84
• /
• 2006

In spite of enormous increase in data generation, its practical usage in the construction sector has not been prevalent enough compared to those of other industries. The author would explore the obstacles against efficient data application in the arena of expenditure forecasting, and suggest a forecasting method by applying Case-based Reasoning (CBR). The newly suggested method in the research, enables project managers to forecast monthly expenditures with less time and effort by retrieving and referring only projects of a similar nature, while filtering out irrelevant cases included in database. Among 99 projects collected, the cost data from 88 projects were processed to establish a new forecasting model. The remaining 10 projects were utilized for the validation of the model. From the comprehensive study, the choice of the numbers of referring projects was investigated in detail. It is concluded that selecting similar projects at 12~19 % out of the whole database will produce a more precise forecasting. The new forecasting model, which suggests the predicted values based on previous projects, is more than just a forecasting methodology; it provides a bridge that enables current data collection techniques to be used within the context of the accumulated information. This will eventually help all the participants in the construction industry to build up the knowledge derived from invaluable experience.

• Journal of Institute of Control, Robotics and Systems
• /
• v.12 no.7
• /
• pp.631-636
• /
• 2006

In this paper, an algorithm of time series function forecasting using GMDH(group method of data handling) algorithm that gives more weight to the recent data is proposed. Traditional methods of GMDH forecasting gives same weights to the old and recent data, but by the point of view that the recent data is more important than the old data to forecast the future, an algorithm that makes the recent data contribute more to training is proposed for more accurate forecasting. The average error rate of electric power demand forecasting by the traditional GMDH algorithm which does not use data weighting algorithm is 0.9862 %, but as the result of applying the data weighting GMDH algorithm proposed in this paper to electric power forecasting demand the average error rate by the algorithm which uses data weighting algorithm and chooses the best data weighting rate is 0.688 %. Accordingly in forecasting the electric power demand by GMDH the proposed method can acquire the reduced error rate of 30.2 % compared to the traditional method.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.65 no.7
• /
• pp.1144-1150
• /
• 2016

Short-term load forecasting is essential to the electricity pricing and stable power system operations. The conventional weekday 24-hour load forecasting algorithms consider the temperature model to forecast maximum load and minimum load. But 24-hour load pattern forecasting models do not consider temperature effects, because hourly temperature forecasts were not present until the latest date. Recently, 3 hour temperature forecast is announced, therefore hourly temperature forecasts can be produced by mathematical techniques such as various interpolation methods. In this paper, a new 24-hour load pattern forecasting method is proposed by using similar day search considering the hourly temperature. The proposed method searches similar day input data based on the anomalous weather features such as continuous temperature drop or rise, which can enhance 24-hour load pattern forecasting performance, because it uses the past days having similar hourly temperature features as input data. In order to verify the effectiveness of the proposed method, it was applied to the case study. The case study results show high accuracy of 24-hour load pattern forecasting.

• Journal of the Korean earth science society
• /
• v.42 no.6
• /
• pp.605-622
• /
• 2021

Seasonal forecasting has numerous socioeconomic benefits because it can be used for disaster mitigation. Therefore, it is necessary to diagnose and improve the seasonal forecast model. Moreover, the model performance is partly related to the ocean model. This study evaluated the hindcast performance in the upper ocean of the Global Seasonal Forecasting System version 5-Global Couple Configuration 2 (GloSea5-GC2) using a multivariable integrated evaluation method. The normalized potential temperature, salinity, zonal and meridional currents, and sea surface height anomalies were evaluated. Model performance was affected by the target month and was found to be better in the Pacific than in the Atlantic. An increase in lead time led to a decrease in overall model performance, along with decreases in interannual variability, pattern similarity, and root mean square vector deviation. Improving the performance for ocean currents is a more critical than enhancing the performance for other evaluated variables. The tropical Pacific showed the best accuracy in the surface layer, but a spring predictability barrier was present. At the depth of 301 m, the north Pacific and tropical Atlantic exhibited the best and worst accuracies, respectively. These findings provide fundamental evidence for the ocean forecasting performance of GloSea5.

• Journal of Intelligence and Information Systems
• /
• v.18 no.4
• /
• pp.43-57
• /
• 2012

To enhance the competitive advantage in a constantly changing business environment, an enterprise management must make the right decision in many business activities based on both internal and external information. Thus, providing accurate information plays a prominent role in management's decision making. Intuitively, historical data can provide a feasible estimate through the forecasting models. Therefore, if the service department can estimate the service quantity for the next period, the service department can then effectively control the inventory of service related resources such as human, parts, and other facilities. In addition, the production department can make load map for improving its product quality. Therefore, obtaining an accurate service forecast most likely appears to be critical to manufacturing companies. Numerous investigations addressing this problem have generally employed statistical methods, such as regression or autoregressive and moving average simulation. However, these methods are only efficient for data with are seasonal or cyclical. If the data are influenced by the special characteristics of product, they are not feasible. In our research, we propose a forecasting framework that predicts service demand of manufacturing organization by combining Case-based reasoning (CBR) and leveraging an unsupervised artificial neural network based clustering analysis (i.e., Self-Organizing Maps; SOM). We believe that this is one of the first attempts at applying unsupervised artificial neural network-based machine-learning techniques in the service forecasting domain. Our proposed approach has several appealing features : (1) We applied CBR and SOM in a new forecasting domain such as service demand forecasting. (2) We proposed our combined approach between CBR and SOM in order to overcome limitations of traditional statistical forecasting methods and We have developed a service forecasting tool based on the proposed approach using an unsupervised artificial neural network and Case-based reasoning. In this research, we conducted an empirical study on a real digital TV manufacturer (i.e., Company A). In addition, we have empirically evaluated the proposed approach and tool using real sales and service related data from digital TV manufacturer. In our empirical experiments, we intend to explore the performance of our proposed service forecasting framework when compared to the performances predicted by other two service forecasting methods; one is traditional CBR based forecasting model and the other is the existing service forecasting model used by Company A. We ran each service forecasting 144 times; each time, input data were randomly sampled for each service forecasting framework. To evaluate accuracy of forecasting results, we used Mean Absolute Percentage Error (MAPE) as primary performance measure in our experiments. We conducted one-way ANOVA test with the 144 measurements of MAPE for three different service forecasting approaches. For example, the F-ratio of MAPE for three different service forecasting approaches is 67.25 and the p-value is 0.000. This means that the difference between the MAPE of the three different service forecasting approaches is significant at the level of 0.000. Since there is a significant difference among the different service forecasting approaches, we conducted Tukey's HSD post hoc test to determine exactly which means of MAPE are significantly different from which other ones. In terms of MAPE, Tukey's HSD post hoc test grouped the three different service forecasting approaches into three different subsets in the following order: our proposed approach > traditional CBR-based service forecasting approach > the existing forecasting approach used by Company A. Consequently, our empirical experiments show that our proposed approach outperformed the traditional CBR based forecasting model and the existing service forecasting model used by Company A. The rest of this paper is organized as follows. Section 2 provides some research background information such as summary of CBR and SOM. Section 3 presents a hybrid service forecasting framework based on Case-based Reasoning and Self-Organizing Maps, while the empirical evaluation results are summarized in Section 4. Conclusion and future research directions are finally discussed in Section 5.

• Management & Information Systems Review
• /
• v.30 no.2
• /
• pp.237-255
• /
• 2011

This empirical study is focused on practical application of Range-Based Volatility which is estimated by opening, high, low, closing price of overall asset. Especially proper forecasting period is what I want to know. There is four useful Range-Based Volatility(RV) such as Parkinson(1980; PK), Garman and Klass(1980; GK) Rogers and Satchell(1991; RS), Yang and Zhang(2008; YZ). So, four RV of KOPSI 200 index during 2000.5.22-2009.9.18 was used for empirical test. The emprirical result as follows. First, the best RV which shows the best forecasting performance is PK volatility among PK, GK, RS, YZ volatility. According to estimating period forcasting performance of RV shows delicate difference. PK has better performance in the period with financial crisis of sub-prime mortgage loan. if not, RS is better. Second, almost result shows better performance on forecasting volatility without sub-prime mortgage loan period. so we can say that forecasting performance is lower when historical volatiltiy is comparatively high. Finally, I find that longer estimating period in AR(1) and MA(1) model can reduce forecasting error. More interesting point is that the result shows rapid decrease form 60 days to 90 days and there is no more after 90 days. So, if we forecast the volatility using Range-Based volaility it is better to estimate with 90 trading period or over 90 days.

• Journal of Environmental Science International
• /
• v.8 no.3
• /
• pp.349-354
• /
• 1999

The transfer function was introduced to establish the prediction method for the DO concentration at the intaking point of Kongju Water Works System. In the mose cases we analyze a single time series without explicitly using information contained in the related time series. In many forecasting situations, other events will systematically influence the series to be forecasted(the dependent variables), and therefore, there is need to go beyond a univariate forecasting model. Thus, we must bulid a forecasting model that incorporates more than one time series and introduces explicitly the dynamic characteristics of the system. Such a model is called a multiple time series model or transfer function model. The purpose of this study is to develop the stochastic stream water quality model for the intaking station of Kongju city waterworks in Keum river system. The performance of the multiplicative ARIMA model and the transfer function noise model were examined through comparisons between the historical and generated monthly dissolved oxygen series. The result reveal that the transfer function noise model lead to the improved accuracy.