Hosseini, Kobra, Keramati, Mohammad Ali, Afshar Kazemi, Mohammad Ali, Fathi, Zadallah. (1404). Designing a model for financial forecasting using the integration of neural networks, Box Jenkins and Holt Winters methodology. هنرهای کاربردی, 16(9), 95-106. doi: 10.22075/ijnaa.2022.27996.3776
Kobra Hosseini; Mohammad Ali Keramati; Mohammad Ali Afshar Kazemi; Zadallah Fathi. "Designing a model for financial forecasting using the integration of neural networks, Box Jenkins and Holt Winters methodology". هنرهای کاربردی, 16, 9, 1404, 95-106. doi: 10.22075/ijnaa.2022.27996.3776
Hosseini, Kobra, Keramati, Mohammad Ali, Afshar Kazemi, Mohammad Ali, Fathi, Zadallah. (1404). 'Designing a model for financial forecasting using the integration of neural networks, Box Jenkins and Holt Winters methodology', هنرهای کاربردی, 16(9), pp. 95-106. doi: 10.22075/ijnaa.2022.27996.3776
Hosseini, Kobra, Keramati, Mohammad Ali, Afshar Kazemi, Mohammad Ali, Fathi, Zadallah. Designing a model for financial forecasting using the integration of neural networks, Box Jenkins and Holt Winters methodology. هنرهای کاربردی, 1404; 16(9): 95-106. doi: 10.22075/ijnaa.2022.27996.3776

Designing a model for financial forecasting using the integration of neural networks, Box Jenkins and Holt Winters methodology

International Journal of Nonlinear Analysis and Applications
مقاله 9، دوره 16، شماره 9، آذر 2025، صفحه 95-106    اصل مقاله (811.26 K)
نوع مقاله: Research Paper
شناسه دیجیتال (DOI): 10.22075/ijnaa.2022.27996.3776
نویسندگان
Kobra Hosseini؛ Mohammad Ali Keramati* ؛ Mohammad Ali Afshar Kazemi؛ Zadallah Fathi
Department of Management, Central Tehran Branch, Islamic Azad University, Tehran, Iran.
چکیده
The present study designs a model for financial forecasting by integrating neural networks. This retrospective comparative study uses the average price data of OPEC oil from 2003 to 2022 to forecast the period from June 2022 to May 2024. To this end, two time-series models (Box-Jenkins and Holt-Winters) were examined, which in the second stage were incorporated into a hybrid model based on artificial neural networks. The neural network model was developed using Matlab, and the Box-Jenkins time-series model was constructed using SPSS and Eviews software. Based on the results of the error analysis of the Box-Jenkins methodology, among the time series processes ARIMA(5,1,5), ARIMA(4,1,5), ARIMA(3,1,5), and ARIMA(5,1,3), the models demonstrated the best accuracy with MSE values of 61.86, 63.21, 63.29, and 63.62, respectively. The accuracy of the Holt-Winters method was not suitable for time-series forecasting due to the nature of the data. Therefore, the best artificial neural network was designed for combining forecasting methods. This neural network included an input layer with 5 neurons, a hidden layer with 5 neurons, and a single-neuron output layer. The network was trained using the Levenberg-Marquardt algorithm and employed a linear sigmoid activation function. The results indicated that the designed hybrid neural network significantly improved the accuracy of the forecasting methods and enhanced the MSE, MAPE, AIC, and BIC indices.
کلیدواژه‌ها
prediction؛ time series؛ ARIMA؛ neural network
مراجع
آمار
تعداد مشاهده مقاله: 1,331
تعداد دریافت فایل اصل مقاله: 604
دانشگاه سمنان
پیوندهای مفید
آمار
تعداد نشریات22
تعداد شماره‌ها718
تعداد مقالات10,319
تعداد مشاهده مقاله72,318,524
تعداد دریافت فایل اصل مقاله64,042,662