• Journal of The Korean Society of Agricultural Engineers
• /
• v.63 no.1
• /
• pp.11-25
• /
• 2021

The objective of this study was to analyze the impact of activation functions on flood forecasting model based on Artificial neural networks (ANNs). The traditional activation functions, the sigmoid and tanh functions, were compared with the functions which have been recently recommended for deep neural networks; the ReLU, leaky ReLU, and ELU functions. The flood forecasting model based on ANNs was designed to predict real-time runoff for 1 to 6-h lead time using the rainfall and runoff data of the past nine hours. The statistical measures such as R2, Nash-Sutcliffe Efficiency (NSE), Root Mean Squared Error (RMSE), the error of peak time (ETp), and the error of peak discharge (EQp) were used to evaluate the model accuracy. The tanh and ELU functions were most accurate with R2=0.97 and RMSE=30.1 (㎥/s) for 1-h lead time and R2=0.56 and RMSE=124.6~124.8 (㎥/s) for 6-h lead time. We also evaluated the learning speed by using the number of epochs that minimizes errors. The sigmoid function had the slowest learning speed due to the 'vanishing gradient problem' and the limited direction of weight update. The learning speed of the ELU function was 1.2 times faster than the tanh function. As a result, the ELU function most effectively improved the accuracy and speed of the ANNs model, so it was determined to be the best activation function for ANNs-based flood forecasting.

• The Korean Journal of Applied Statistics
• /
• v.26 no.2
• /
• pp.349-360
• /
• 2013

Forecasting the daily peak load for electricity demand is an important issue for future power plants and power management. We first introduce several time series models to predict the peak load for electricity demand and then compare the performance of models under the RMSE(root mean squared error) and MAPE(mean absolute percentage error) criteria.

• Communications for Statistical Applications and Methods
• /
• v.30 no.3
• /
• pp.273-289
• /
• 2023

In this paper, we develop a new time series model for predicting IPO (initial public offering) data with non-negative integer value. The proposed model is based on integer-valued autoregressive (INAR) model with a Poisson thinning operator. Just as the heterogeneous autoregressive (HAR) model with daily, weekly and monthly averages in a form of cascade, the integer-valued heterogeneous autoregressive (INHAR) model is considered to reflect efficiently the long memory. The parameters of the INHAR model are estimated using the conditional least squares estimate and Yule-Walker estimate. Through simulations, bias and standard error are calculated to compare the performance of the estimates. Effects of model fitting to the Korea's IPO are evaluated using performance measures such as mean square error (MAE), root mean square error (RMSE), mean absolute percentage error (MAPE) etc. The results show that INHAR model provides better performance than traditional INAR model. The empirical analysis of the Korea's IPO indicates that our proposed model is efficient in forecasting monthly IPO volumes.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1993.10a
• /
• pp.134-152
• /
• 1993

Many business forecasting problems are characterized by infrequent occurences, a large number of variables, presence of error, and great complexity. Because no forecasting models and tools are effective in handing these problems, managers often use the outcomes of past analogous cases to predict the outcome of the current one. They (1) observe significant attributes in describing a case, (2) identify the past cases similar in these attributes to the current case, and (3) predict the outcome of the current case based on those of the analogous cases identified through some mental simulation and adjustment. This process of forecasting can be termed forecasting-by-analogy. In spite of fairly frequent use of this forecasting process in practice, however, if has not been recognized as a primary forecasting tool, nor applied on a regular basis. In this paper, by automatizing this process using computer models, we develop a case-based forecasting system(CBFS), which identifies relevant cases and applies their outcomes to generate a forecast. We demonstrate the effectiveness of the CBFS in terms of its accuracy in predicting the outcome of the current problem based on the similar cases identified. We compare the forecasting accuracy of the CBFS with that of regression models developed by stepwise procedure under varied simulated problem conditions. The CBFS outperforms regression models in most comparisons. The CBFS could be used as an effective forecasting tool.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.19 no.2
• /
• pp.199-215
• /
• 1994

Many business forecasting problems are characterized by infrequent occurrences, a large number of variables, presence of error, and great complexity. Because no forecasting models and tools are effective in handing these problems, managers often use the outcomes of past analogous cases to predict the outcome of the current one. They (1) observe significant attributes in describing a case, (2) identify the past cases similar in these attributes to the current case, and (3) predict the outcome of the current case based on those of the analogous cases identified through some mental simulation and adjustment. This process of forecasting can be termed forecasting-by-analogy. In spite of fairly frequent use of this forecasting process in practice, however, it has not been recognized as a primary forecasting tool, nor applied on a regular basis. In this paper, by automatizing this process using computer models, we develop a case-based forecasting system (CBFS), which identifies relevant cases and applies their coutcomes to generate a forecast. We demonstrate the effectiveness of the CBFS in terms of its accuracy in predicting the outcome of the current problem based on the similar cases identified. We compare the forecasting accuracy of the CBFS with that of regression models developed by stepwise procedure under varied simulated problem conditions. The CBFS outperforms regression models in most comparisons. The CBFS could be used as an effective forecasting tool.

• The Korean Journal of Applied Statistics
• /
• v.35 no.3
• /
• pp.385-393
• /
• 2022

In this paper, we consider univariate time series models that are well known in the field of forecasting and we study on forecasting performance for their simple combinations. The univariate time series models include exponential smoothing methods and ARIMA (autoregressive integrated moving average) models, their extended models, and non-seasonal and seasonal random walk models, which is frequently used as benchmark models for forecasting. The median and mean are simply used for the combination method, and the data set used for performance evaluation is M3-competition data composed of 3,003 various time series data. As results of evaluating the performance by sMAPE (symmetric mean absolute percentage error) and MASE (mean absolute scaled error), we assure that the simple combinations of the univariate models perform very well in the M3-competition dataset.

• The Korean Journal of Applied Statistics
• /
• v.23 no.2
• /
• pp.249-261
• /
• 2010

Long-term forecasting of seasonal time series is critical in many applications such as planning business strategies and resolving possible problems of a business company. Unlike the traditional approach that depends solely on dynamic models, Li and Hinich (2002) introduced a combination of stochastic dynamic modeling with filter bank approach for forecasting seasonal patterns using highly coherent(High-C) waveforms. We modify the filter selection and forecasting procedure on wavelet domain to be more feasible and compare the resulting predictor with one that obtained from the wavelet variance estimation method. An improvement over other seasonal pattern extraction and forecasting methods based on such as wavelet scalogram, Holt-Winters, and seasonal autoregressive integrated moving average(SARIMA) is shown in terms of the prediction error. The performance of the proposed method is illustrated by a simulation study and an application to the real stock price data.

• Proceedings of the Korea Association of Information Systems Conference
• /
• 2000.11a
• /
• pp.36-44
• /
• 2000

The purpose of this study is to introduce a more efficient forecasting technique, which could help result the reduction of cost in removing the waste of airline in-flight meals. We will use a neural network approach known to many researchers as the “Outstanding Forecasting Technique”. We employed a multi-layer perceptron neural network using a backpropagation algorithm. We also suggested using other related information to improve the forecasting performances of neural networks. We divided the data into three sets, which are training data set, cross validation data set, and test data set. Time lag variables are still employed in our model according to the general view of time series forecasting. We measured the accuracy of our model by “Mean Square Error”(MSE). The suggested model proved most excellent in serving economy class in-flight meals. Forecasting the exact amount of meals needed for each airline could reduce the waste of meals and therefore, lead to the reduction of cost. Better yet, it could enhance the cost competition of each airline, keep the schedules on time, and lead to better service.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.43 no.5
• /
• pp.730-738
• /
• 1994

The major advantage of the short-term load forecasting technique using general exponential smoothing is high accuracy and operational simplicity, but it makes large forecasting error when the load changes repidly. The paper has presented new technique to improve those shortcomings, and according to forecasted the technique proved to be valid for two years. The structure of load model is time function which consists of daily-and temperature-deviation component. The average of standard percentage erro in daily forecasting for two years was 2.02%, and this forecasting technique has improved standard erro by 0.46%. As relative coefficient for daily and seasonal forecasting is 0.95 or more, this technique proved to be valid.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.45 no.4
• /
• pp.511-516
• /
• 1996

In this paper, weekly maximum power demand forecasting method in consideration of temperature estimation using a time series model was presented. The method removing weekly, seasonal variations on the load and irregularities variation due to unknown factor was presented. The forecasting model that represent the relations between load and temperature which get a numeral expected temperature based on the past 30 years(1961~1990) temperature was constructed. Effect of holiday was removed by using a weekday change ratio, and irregularities variation was removed by using an autoregressive model. The results of load forecasting show the ability of the method in forecasting with good accuracy without suffering from the effect of seasons and holidays. Percentage error load forecasting of all seasons except summer was obtained below 2 percentage. (author). refs., figs., tabs.