Combining seasonal time series ARIMA method and neural networks with genetic algorithms for predicting the production value of the mechanical industry in Taiwan

Abstract

Supplying industrial firms with an accurate method of forecasting the production value of the mechanical industry to facilitate decision makers in precise planning is highly desirable. Numerous methods, including the autoregressive integrated-moving average (ARIMA) model and artificial neural networks can make accurate forecasts based on historical data. The seasonal ARIMA (SARIMA) model and artificial neural networks can also handle data involving trends and seasonality. Although neural networks can make predictions, deciding the most appropriate input data, network structure and learning parameters are difficult. Therefore, this article presents a hybrid forecasting method that combines the SARIMA model and neural networks with genetic algorithms. Analytical results generated by the SARIMA model are inputted as the input data of a neural network. Subsequently, the number of neurons in the hidden layer and the number of learning parameters of the neural network architecture are globally optimized using genetic algorithms. This model is subsequently adopted to forecast seasonal time series data of the production value of the mechanical industry in Taiwan. The results presented here provide a valuable reference for decision makers in industry.

SARIMA model-based forecasting required number of COVID-19 vaccines globally and empirical analysis of peoples’ view towards the vaccines

“…The SARIMA method has seven parameters. The first three trend parameters represented as describes the non-seasonal part and four of the seasonal parameters represented as deals with the seasonal part of the model [45] , [46] . The mathematical formula of the SARIMA model is shown in Eq.…”

Improving Hybrid Models For Precipitation Forecasting By Combining Nonlinear Machine Learning Methods

“…The purpose of this study was to improve the accuracy of the traditional hybrid model with one linear and one nonlinear in their structures. To achieve this, we 1) properly modeled the linear component of time series using ARIMA or SARIMA; 2) precisely selected the structures of the two configurations with antecedent residual subseries; and with antecedent observations and residual subseries and linear model simulations; 3) properly selected a model to represent residual subseries using an artificial intelligence technique (Liang, 2009;Zheng, 2003;Chen, et al, 2010;Moeeni et al, 2017); and 4) accurately estimated parameters of the selected model. The average RMSE, MAE, UI, and UII decreased from configuration one to two in GEP model for Tabriz and Rasht time series by88% and 56%, respectively; In this study, the performance of SVR and GMDH was better than GEP model.…”

“…A combination of SARIMA and ANN models was used to forecast monthly inflow, and the combined model had a high coefficient of determination relative to these two composing models (Moeeni et al, 2017). Another hybrid model was developed based on ARIMA coupled with ANN using GA to forecast the production value of the mechanical industry (Liang 2009). GA was used to optimize the ANN parameters, such as number of hidden neurons.…”

ARIMA models for predicting the end of COVID-19 pandemic and the risk of second rebound

“…The non-seasonal ARIMA model ( p , d , q ) is vital in building pure seasonal SARIMA model, whereby the term ( p , d , q ) presents the non-seasonal part of the model and describes the seasonal part of the model [ 30 , 31 ]. The mathematical description of the model is presented as shown in Eq.…”