Title The Agricultural Product Wholesale Price Index Forecasting Model Based on LSSVR Optimized by PSO
Authors 李崇光
Issue Date 2012
Publisher Journal of Convergence Information Technology
Keywords Agricultural product wholesale price index (APWPI)
Forecasting
LSSVR
PSO
Citation Journal of Convergence Information Technology, 2012,7(17): 531 -539.
Abstract In recent years, the prices of some agricultural products in China have gone through sudden fluctuations between low and high. Especially, the fluctuation of a few agricultural products' prices between low and high level has even become a hot topic of public opinion. The price fluctuations of agricultural products, on the one hand, play a positive impact on the farmers' income and enthusiasm for production; on the other hand, it is related to people's daily lives and vital interests. Therefore, it is of great significance to correctly understand the price fluctuations of agricultural products, to accurately predict the prices of agricultural products, and to promote the formation of a reasonable price level of the agricultural products. In this paper, a forecasting method based on particle swarm optimization and least squares support vector regression(PSO-LSSVR) is proposed, with the quarterly data of agricultural product wholesale price index (APWPI) between 2009 and 2011 as the sample. Specifically in this sample, the data between the first quarter of 2009 and the first quarter of 2011 is selected as the training sample, and the forecasting sample is the data ranging from the 2nd to 4th quarter in 2011. The simulation results show that the mean absolute percentage error (MAPE) of PSOLSSVR is obviously lower than that of back-propagation neural network (BPN), which indicates that the method based on PSO-LSSVR has higher accuracy to forecast the APWPI. Finally, some corresponding suggestions and strategies are put forward to stabilize the long-term mechanism of agricultural products' prices and to inhibit the rise of APWPI.
Appears in Collections: 校领导

Original Search


Files in This Work
There are no files associated with this item.

Google Scholar™






License: See PKU IR operational policies.