英语翻译3.6.Evaluations and comparisonsIn order to compare various models of the prediction error,thisarticle by MAPE measure in-sample and out-of-sample dataaccuracy of the forecast,and the results are shown in Table 4.The in-sample data are use

英语翻译3.6.Evaluations and comparisonsIn order to compare various models of the prediction error,thisarticle by MAPE measure in-sample and out-of-sample dataaccuracy of the forecast,and the results are shown in Table 4.The in-sample data are use
英语翻译
3.6.Evaluations and comparisons
In order to compare various models of the prediction error,this
article by MAPE measure in-sample and out-of-sample data
accuracy of the forecast,and the results are shown in Table 4.
The in-sample data are used to establish the grey forecasting model.
The out-of-sample data are used to evaluate the forecasting
performance.
Table 4 shows that GM(1,1),GM(1,6),AGAGM(1,6) and
GAGM(1,6) in the in-sample (1993–2003) of the MAPE values are
0.146,0.2875,0.2378 and 0.2009,respectively.
With regards to the out-of sample (2004–2005) results of the
forecast,GM(1,1),GM(1,6),AGAGM(1,6) and GAGM(1,6) MAPE values
are 0.0828,0.0804,0.0464 and 0.0076,respectively.
Considering all the empirical results for out-of-sample,the
GAGM(1,6) model had the smallest MAPE (0.76%) with an excellent
forecasting power (Lewis,1982).Witt and Witt (1992) and Law
(2000) mentioned that the MAPE value (0.76%) is very low,with
excellent forecast accuracy.In other words,this article proposed
that AGAGM(1,6) and GAGM(1,6) can effectively improve the original
grey model forecast accuracy.

英语翻译3.6.Evaluations and comparisonsIn order to compare various models of the prediction error,thisarticle by MAPE measure in-sample and out-of-sample dataaccuracy of the forecast,and the results are shown in Table 4.The in-sample data are use
3.6.评价和比较为了比较各种模型的预测误差,这本文的结果措施的样本和样本数据预测的准确性,和的结果见表4.使用的样本数据建立的灰色预测模型.样本外数据是用来评估预测性能.表4显示,通用汽车（1 ,1）,通用汽车（1 ,6）,agagm(不知）（1）和伽格美（1 ,6）的样本（1993–2003）的平均相对误差值0.146,0.2875,0.2378和0.2009,分别为.至于外的样本（2004–2005）结果的预测,通用汽车（1 ,1）,通用汽车（1 ,6）,agagm（1 ,6）和伽格美（1 ,6）结果值0.0828,0.0804,0.0464和0.0076,分别为.考虑到所有的研究样本,这伽格美（1）模型有最小的结果（0.76%）与一个很好的预测能力（路易斯,1982）.维特,维特（1992）和法（2000）提到,平均相对误差值（0.76%）很低,与良好的预测精度.换句话说,本文提出这agagm（1 ,6）和伽格美（1 ,6）可以有效地提高原灰色模型的预测精度.