水文变异的滑动F识别与检验方法
DOI:
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

P333.6

基金项目:

全国水资源综合规划专题研究项目


Slide F test of Change-Point Analysis
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    针对传统F检验法只能用于已知变异点的检验,而无法用于变异点的识别问题,本文提出了滑动F识别与检验法。该法首先利用传统F检验法对序列逐点进行检验,找出满足F计算值大于临界值的所有可能变异点,然后选择F计算值达到最大值的那一点作为最可能变异点。利用该法对潮白河水资源分区45年年径流量序列进行了分析,并从物理成因角度对该检验结果进行了确认。结果表明滑动F 识别与检验法具有较好的检验效果,检验结果可信度较高。

    Abstract:

    On account of the influence of human activities and climate change, the hydrologic series may change in time and space. Usually, the statistical method of F test is used to test change-point, but it can't be used to check out change-point. A slide F test of change-point analysis method is presented in this paper. Firstly, it test all of the points in the time series to get all possible change-points. Then, it chooses the point whose calculated F value is the maximum F as the probable change-point. The method has been used to analyze Chaobaihe River's 45 years annual runoff with the results from physical cause being confirmed, which shows that slide F test is better than traditional F test. It can be used to test and check out change-point. It is more effective in change-point analysis and the result is more reliable.

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2005-04-28
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期:
  • 出版日期: