A new approach for forecast sales growth in automobile industry

In this study, a novel hybrid model is used for automobile sales forecasting based on the prediction results of time series forecasting. In order to improve the prediction or forecasting capability, the study integrates an Adaptive multiplicative triple exponential smoothing Holt-Winters (AHW) method and Backpropagation Neural Networks (BPNNs). The former method sets the weights required for Backpropagation in Neural Networks to improve the forecasting or prediction ability in the field of automobile sales industry. The proposed technique uses automobile sales data of recent years and prediction outcomes of time series to train the proposed forecasting model. The validation is carried out using real-world automobile sales data for both training and testing purposes. The result of proposed method outperforms the other neural network forecasting model in terms of automobile sales forecasting. © 2020 IJSTR.