中文摘要：

分别利用Leontief逆矩阵和Ghosh逆矩阵构建增长态归因矩阵，提出完全需求拉动作用和完全供给推动作用的评价模型。实证分析表明：增长态归因矩阵法可有效地辨识产业动态关联，有助于选择产业关键部门；拉动作用增强的产业主要依赖间接拉动作用，而其他产业则主要依赖直接拉动作用；就推动作用而言也能得到相同结论。

英文摘要：

This paper establishes growing causative matrixesrespectively by using Leontief inverse matrix and Gosh inverse matrix, and the evaluation model of pull effect of full demand and push effect of supply are provided respectively . Result shows that：the method of growing causative matrix is more effective to identify industrial sectors＇ dynamic relationships, and should be contribute to select ＇ key＇ industrial sectors; the industrial sectors whose pull effect are enhanced mainly depend on indirect pull effect, but the other sectors mainly depend on direct pull effect. Similarly, as for the push effect, the same conclusion also can be obtained. The suggestions in it is that the industrial structure adjustment in China should base on the demand pulling, and should give priority to the industrial sectors whose pulling effect and pushing effect are all enhanced, thus the industry structure can be optimized and the healthy growth of the industry can be promoted through the pulling effect and pushing effect.