本文首先提出了以前文献对DGM(1,1)与GM(1,1)模型关系的证明错误,否定了以前文献的一些观点.第二,证明了DOOM(1,1)模型的MAPE值与边值x(1)(m)(1≤m≤n)无关.第三,发现DOOM(1,1)模型的MAPE值变劣现象并分析其原因.第四,用最小一乘法,构建了边值优化的DGOMII(1,1)模型,其效果优于DGOM(1,1)模型.最后,将DGM(1,1)及DGOMII(1,1)模型应用于技术创新领域中.
This paper firstly put forward the proof errors on the relationship between model DGM(1,1) and GM(1,1) and denied some viewpoints in the previous literatures; Secondly, the MAPE value of model DGOM(1,1) was proved to be independent of the initial value x(1)(m) (1 ≤m ≤ n); Thirdly, the deteri- oration phenomenon of the MAPE value of model DGOM(1,1) was found and the reason was analyzed; Fourthly, its performance of model DGOMII(1,1) based on the initial value optimization by least absolution deviation was better than that of DGOM(1,1); Finally, model DGM(1,1) and DGOMII(1,1) were applied to the field of technology innovation.