基于时间序列的黑龙江省粮食产量预测

Prediction of Crop Yields in Heilongjiang Province Based on Time Series

应用时间序列分析方法,建立粮食生产动态模型,利用黑龙江省1980～2006年粮食总产量的数据进行了粮食产量的预测。将预测数据实际数据对比,结果表明:抛物线型趋势线平均百分误差为9.60,平均绝对误差为215.80,平均偏差平方和为70 281.90是准确度最高的。而指数型趋势线预测值在最近4年内最接近实际产量,其精度分别达到98.32%、97.31%、96.35%和94.02%。

The crop production dynamic model is established by time series analysis method.With which the 1980~2006 year crop production data of Heilongjiang Province is analyzed.The result is compared with the actual data.The result shows that the parabolic trend line has the highest accuracy. Its average percentage error is 9.60, average absolute error is 215.80, and average deviation square is 70 281.90.The exponential trend line predicts in nearly 4 years closest to the actual production of Heilongjiang Province. The accuracy reaches to 98.32%, 97.31%, 96.35% and 94.02%.