Name
Title | ||
|---|---|---|
Livestock Product Price Forecasting Method Based on Heterogeneous GRU Neural Network and Energy Decomposition |
Abstract | ||
|---|---|---|
The characteristics exhibited by livestock product price fluctuation should be characterized, and the trend of price fluctuation should be forecasted in times, which are critical to developing the animal husbandry market. The current mainstream forecasting methods have limited ability to deal with multisource, multi-modal and heterogeneous input data. Moreover, the current mainstream forecasting methods ignored the impact of static non-time series such as variety, growth cycle, latitude and longitude on the price fluctuation of livestock products, so it can not accurately predict the livestock product price and fluctuation trend. To introduced the influence of static information on price fluctuation into the forecasting process while improved the ability to process heterogeneous data. A novel price forecasting method was proposed by complying with GRU neural network and the principle of energy decomposition. First, to acquire the price fluctuation at different frequencies, this study proposed a variation mode decomposition method based on actual signal energy (AE-VMD) and a multi-scale adaptive Lempel-Ziv complexity calculation method (MA-LZ). Second, to preserve the information of multimodal data, this study developed a heterogeneous GRU neural network (AH-GRU) in accordance with attention mechanism. Lastly, the effect of static information on price fluctuations was introduced in the forecasting, and the final forecasting result was outputted via the dense layer. The experimental results showed that the RMSE of the forecasting method for beef, mutton and pork prices were 0.726, 0.535 and 0.738, the MAE were 0.607, 0.412 and 0.621, and the R-2 were 0.954, 0.995 and 0.991, respectively. The trend forecasting statistics D-stat were 87.097, 90.909 and 90.566, the correct upward trend (CP) were 68.805, 41.818 and 42.875, and the correct downward trend (CD) were 63.326, 59.091 and 47.619, respectively. The proposed method outperformed the mainstream livestock product price forecasting method in forecasting accuracy, trend forecasting and method convergence. |
Year | DOI | Venue |
|---|---|---|
2021 | 10.1109/ACCESS.2021.3128960 | IEEE ACCESS |
Keywords | DocType | Volume |
Heterogeneous GRU neural network, price forecasting, energy decomposition | Journal | 9 |
ISSN | Citations | PageRank |
2169-3536 | 0 | 0.34 |
References | Authors | |
0 | 6 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Keqiang Li | 1 | 583 | 52.39 |
| Nan Shen | 2 | 0 | 0.34 |
| Yan Kang | 3 | 0 | 0.34 |
| Hong Chen | 4 | 280 | 56.04 |
| Yuqing Wang | 5 | 0 | 0.34 |
| Shiqian He | 6 | 0 | 0.34 |