Abstract
In this paper, we focus on the problem of dynamically monitoring agricultural economy, which is a key problem in economic development of the national economy. Firstly, we describe the structure of the dynamically monitoring agricultural economy system, which is made up of four steps: 1) collecting the agricultural economy data, 2) estimating parameters of the proposed monitoring model, 3) obtaining the agricultural economy prediction results, and 4) calculating the error rate. As the time series data of agricultural economy includes both complex linear and nonlinear patterns, it is not easy to promote the prediction accuracy rates using only linear or neural network models. Therefore, the decision support process for agricultural economy dynamically monitoring is implemented via combining the linear regression model and the neural network model. Thirdly, experiments are designed to make performance evaluation, and experimental results show that our algorithm can effectively lower the error rate of agricultural economy monitoring both with time and region changing.
