摘要
根据城市用水量的影响因素及特点,针对传统的线性回归模型误差较大的缺点,基于核估计与局部线性估计理论,建立了城市日用水量的非参数回归预测模型。经西安市实例验证表明,相对于线性回归模型而言,多元非参数回归模型能够很好地解决城市日用水量预测这一问题,预测精度较高,可以满足供水系统调度的实际需要。
According to the effecting factor and the characteristic of urban daily water and the error of line model,we set up the nonparametric multi regressive model of daily water demand. The regression function m（ · ） is estimated by the kernel estimation and local linear estimation. By the experiment of Xi＇an city, it was proved that the nonparametric multi regression model can obtain optimal result and meet the practical requirement of water supply dispatch system.
出处
《青岛科技大学学报（自然科学版）》
CAS
2007年第1期65-68,共4页
Journal of Qingdao University of Science and Technology:Natural Science Edition
关键词
非参数回归模型
日用水量
核估计
局部线性估计
nonparametric regression
water demand
local linear estimation
kernel estimation
