基于ARMA模型的四川省农村居民收入趋势预测

基于1978-2010年四川省农村居民人均纯收入的时间序列，先后通过平稳化检验及处理、自相关与偏相关分析，并结合SC准则定阶，建立模型为 ；然后确定模型参数并对模型的残差序列进行白噪声检验，检验通过，满足预测的要求；接着，运用该模型对四川省农村居民人均纯收入进行预测，结果显示平均绝对误差率为1.81%，其中2010年误差率为1.21%，精度较高；最后对2011-2013年进行短期预测，数据表明农村居民人均收入将继续增长。

The forecasting trend of rural residents income in Sichuan province based on ARMA model

Based on the time series of rural residents per net income in Sichuan Province from 1978 to 2010, through stationary test and disposal, auto correlation and partial correlation analysis, and combing SC criterion, the model of ARMA （5,5） was established. After model parameters were estimated, white noise test about residual series of the model was done. It was showed that results pass test and meet forecast demand. Then rural residents per net incomes in Sichuan Province were estimated with the model, and mean absolute per error was 1.81%, error ratio in 2010 was 1.21%, therefore the precision was higher. Finally rural residents per net incomes in 2011 to 2013 were forecasted, it was demonstrated that the incomes would kept on rise and the increase range was downtrend.