Corn protein powder price forecasting based on parameter optimization VMD and XGBoost algorithm
The stabilization of corn protein powder price is of great significance to the sustainable development of the feed industry and national food security,but its price series is characterized by non-smooth and non-linear features,which makes it difficult to predict accurately.This study aims to construct a corn protein powder price prediction model based on the XGBoost algorithm.First,the whale algorithm(WOA)is used to optimize the K-value and penalty parameter of the variational mode decomposition(VMD)to adaptively decompose the original price series and reduce the data noise.Second,the Pearson feature-screened variables are used as inputs to the Extreme Gradient Boosting Tree(XGBoost)model for training and testing.Finally,ten-fold cross-validation and learning curves are used to test the model performance and analyze the nonlinear effects of key influencing factors in conjunction with the SHAP model.The study showed that the previous period's soybean meal futures price had a significant positive effect on the current period's corn protein meal price volatility.The results show that the XGBoost model optimized by Bayesian algorithm(BO)has a better predictive performance than the benchmark model.
XGBoost algorithmprice predictioncorn protein powdervariational modal decompositionSHAP modelBayesian optimization
国家科技期刊平台
NSTL
万方数据
