The corn output in a time series prediction model

This paper selected the raw data of the corn output of Dehui City, Jilin Province from 1990 to 2000, through data cleansing, data conversion and data integration technologies obtains time series data set, choosing the appropriate time series methods ARIMA(Autoregressive Integrated Moving Average) to confirm the corn output in a time series prediction model. The experimental results show that comparing the actual output and the prediction achieved by the model of corn output from 2001 to 2003, the error is very small; the relative error can be controlled within 5%. This proves that the ARIMA(2,2,1) model can fairly predict the developing trend of the corn output in this region, and the result of the prediction can provide very important theory evidence for the agricultural production management department to make decisions.