统计研究 ›› 2009, Vol. 26 ›› Issue (6): 72-78.

• 论文 • 上一篇    下一篇

基于因子模型的快速灾情统计指标体系研究

四川省统计局《地震快速灾情统计体系研究》课题组   

  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2009-06-15 发布日期:2009-06-15

Rapid Disaster Statistics Index System Research based on the Factor Model

  • Received:1900-01-01 Revised:1900-01-01 Online:2009-06-15 Published:2009-06-15

摘要: 鉴于我国尚无完整的灾害损失统计制度和地震快速灾情统计指标体系,四川汶川8.0级地震对我国灾害损失统计制度方法提出了新的挑战。本文从完善统计制度和方法的角度出发,基于因子分析模型,通过模型的不断改进,试图对众多灾损指标进行多步降维处理,在尽量保持信息量丢失最少的前提下,反复比较多个模型间的有效性和一致性,筛选出稳定的、有效的重要灾损指标,构建了中国地震快速灾情统计指标体系。并以四川省138个受灾县(市、区)的数据对所建地震灾情快速反应统计指标体系进行了实证分析。结果表明:所建中国地震灾情快速反应指标体系具有简便易行、代表性强、时效性高的优势特征,与中国地震局公布的地震烈度图保持了一致性。

关键词: 因子模型, 快速灾情统计, 指标体系

Abstract: From the perspective of the statistical system and methods improvement, we attempted to reduce the dimension of numerous damage indicators by multi-step based on the factor analysis model. In the loss at least as far as possible to maintain the amount of information under the premise, we selected a stable and effective important indicators system of damage on repeated comparison between the effectiveness and consistency of various models, finally we constructed China’s rapid earthquake disaster statistics index system through the model of continuous improvement.