• 응용통계연구
• /
• 제32권1호
• /
• pp.69-81
• /
• 2019

본 논문에서는 딥러닝을 이용한 이변량 장기종속시계열(long-range dependent time series) 예측을 고려하였다. 시계열 데이터 예측에 적합한 LSTM(long short-term memory) 네트워크를 이용하여 이변량 장기종속시계열을 예측하고 이를 이변량 FARIMA(fractional ARIMA) 모형인 FIVARMA 모형과 VARFIMA 모형과의 예측 성능을 실증 자료 분석을 통해 비교하였다. 실증 자료로는 기능적 자기공명 영상(fMRI) 및 일일 실현 변동성(daily realized volatility) 자료를 이용하였으며 표본외 예측(out-of sample forecasting) 오차 비교를 통해 예측 성능을 측정하였다. 그 결과, FIVARMA 모형과 VARFIMA 모형의 예측값에는 미묘한 차이가 존재하며, LSTM 네트워크의 경우 초매개변수 선택으로 복잡해 보이지만 계산적으로 더 안정되면서 예측 성능도 모수적 장기종속시계열과 뒤지지 않은 좋은 예측 성능을 보였다.

• 산업공학
• /
• 제16권4호
• /
• pp.400-410
• /
• 2003

Most of the demand forecasting which have been studied is about long-term time series over 15 years demand forecasting. In this paper, we set up the most optimal ARIMA model for the short-term time series demand forecasting and suggest demand forecasting system for short-term time series by appraising suitability and predictability. We are going to use the univariate ARIMA model in parallel with the bivariate transfer function model to improve the accuracy of forecasting. We also analyze the effect of advertisement cost, scale of branch stores, and number of clerk on the establishment of marketing policy by applying statistical methods. After then we are going to show you customer's needs, which are number of buying products. We have applied this method to forecast the annual sales of refrigerator in four branch stores of A company.

• International Journal of Advanced Culture Technology
• /
• 제11권4호
• /
• pp.393-405
• /
• 2023

Accurate prediction of wind speed and power is vital for enhancing the efficiency of wind energy systems. Numerous solutions have been implemented to date, demonstrating their potential to improve forecasting. Among these, deep learning is perceived as a revolutionary approach in the field. However, despite their effectiveness, the noise present in the collected data remains a significant challenge. This noise has the potential to diminish the performance of these algorithms, leading to inaccurate predictions. In response to this, this study explores a novel feature engineering approach. This approach involves altering the data input shape in both Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) and Autoregressive models for various forecasting horizons. The results reveal substantial enhancements in model resilience against noise resulting from step increases in data. The approach could achieve an impressive 83% accuracy in predicting unseen data up to the 24th steps. Furthermore, this method consistently provides high accuracy for short, mid, and long-term forecasts, outperforming the performance of individual models. These findings pave the way for further research on noise reduction strategies at different forecasting horizons through shape-wise feature engineering.