以隶属度为权重的加权回归模型及在水文水资源中的应用
DOI:
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

P333.9

基金项目:


Weighted Regression Model and Its Application with Membership Function as Weight
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:

    概率统计方法在水文水资源中的应用已有很长历史,常用的是线性回归模型。然而,等权线性回归模型的误差不仅与同步观测资料的长度及残差平方和有关,还随自变量与其均值的变幅加大而增大。因此,线性回归模型一般不应该用来外延。基于加权回归模型的基本理论,以隶属度为权重建立了加权回归模型,并证明了加权回归估计值的无偏性。通过实例分析计算,说明了方法的可行性。

    Abstract:

    It is long time to apply hydrologic statistics in series extension. The method to be used is a linear regression model. However, the error of liner regression model is not only concerened with the length of synchronous series and square sum of residue difference, but also increases with distance between independent variable to its mean. Generally, the regression model is not applied to series extension. The weighted regression model is established based on linear regression model with membership function as weight and its unbiasedness being proved. The model has used for series extension in ZIJINGUAN basin and yields satisfactory results.

    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:2005-05-09
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期:
  • 出版日期: