An optimized hybrid model based on artificial intelligence for grape price forecasting

Abstract

Purpose

The purpose of this paper is to propose an optimized hybrid model based on artificial intelligence methods, use the method of time series forecasting, to deal with the price prediction issue of China’s table grape.

Design/methodology/approach

The approaches follows the framework of “decomposition and ensemble,” using ensemble empirical mode decomposition (EEMD) to optimize the conventional price forecasting methods, and, integrating the multiple linear regression and support vector machine to build a hybrid model which could be applied in solving price series predicting problems.

Findings

The proposed EEMD-ADD optimized hybrid model is validated to be considered satisfactory in a case of China’ grape price forecasting in terms of its statistical measures and prediction performance.

Practical implications

This study would resolve the difficulties in grape price forecasting and provides an adaptive strategy for other agricultural economic predicting problems as well.

Originality/value

The paper fills the vacancy of concerning researches, proposes an optimized hybrid model integrating both classical econometric and artificial intelligence models to forecast price using time series method.

Citation

Chu, X., Li, Y., Tian, D., Feng, J. and Mu, W. (2019), "An optimized hybrid model based on artificial intelligence for grape price forecasting", British Food Journal, Vol. 121 No. 12, pp. 3247-3265. https://doi.org/10.1108/BFJ-06-2019-0390