A Forecast Combination Framework with Multi‐Time Scale for Livestock Products’ Price Forecasting

Abstract

China’s livestock market has experienced exceptionally severe price fluctuations over the past few years. In this paper, based on the well-established idea of “forecast combination,” a forecast combination framework with different time scales is proposed to improve the forecast accuracy for livestock products. Specifically, we combine the forecasts from multi-time scale, i.e., the short-term forecast and the long-term forecast. Forecasts derived from multi-time scale introduce complementary information about the dynamics of price movements, thus increasing the diversities within the modeling process. Moreover, we investigate a total of ten combination methods with different weighting schemes, including linear and nonlinear combination. The empirical results show that (i) forecast performance can be remarkably improved with this novel combination idea, and short-term forecast model is more suitable for the products with a relatively high volatility, e.g., mutton and beef; (ii) geometric mean, which provides a nonlinear combination, is the most effective one among all the combination methods; and (iii) variance-based weighting scheme can yield a superior result compared to the best individual forecast, especially for the products such as egg and beef.

Agricultural Product Price Forecasting Methods: A Review

“…Some researchers have found that time series data of different time scales contain different amounts of information, and combining data from different time scales can enhance prediction performance. Ling et al [68] were the first to attempt the combination of forecasts from multiple time scales, proposing a novel multi-time scale combination strategy for forecasting Chinese livestock product prices. The research results indicate that adopting this new combination approach can significantly improve prediction performance.…”

A Poisson-Regression, Support Vector Machine and Grey Prediction Based Combined Forecasting Model Proposal: A Case Study in Distribution Business

“…The combination methods used are as follows. [35] In Equal Weights (EW) approach, equal weights are assigned to each forecast. We chose w1= 1/3, w2 = 1/3, and w3 = 1/3.…”

Multi-step prediction of carbon emissions based on a secondary decomposition framework coupled with stacking ensemble strategy

“…Although hybrid models perform effectively in the field of carbon emission prediction, the characteristics of carbon emission data vary widely in different countries and regions, and no single hybrid model can be considered suitable for all prediction scenarios (Jiang 2021 ). Combined prediction utilizes several individual models of varying ability to capture nonlinearities and perform differently on various datasets, to obtain better generalization performance (Ling et al 2019 ). Allende and Valle ( 2016 ) considers that the use of suitable ensemble methods contributes to improve the overall prediction accuracy, and the commonly used integration methods are simple averaging (Graefe et al 2014 ) and weighted averaging (Kourentzes et al 2019 ).…”