西安市用水需求量的多元线性回归预测

本文分析了西安市用水需求量的影响因素,选取了6个变量,收集了西安市2004年-2018年的全年用水总量和6个变量的数据建立了用水需求量与6个变量的多元线性回归模型。经检验,模型具有较高的拟合优度和整体显著性,但有个别变量的系数的线性显著性较低,为此运用SPSS建立了多元线性逐步回归模型。经检验,模型可用来对西安市的用水量进行模拟及预测。最后运用所建立的多元线性逐步回归模型对西安市2019-2022全年用水总量进行了预测。

Multiple Linear Regression Prediction of Water Demand in Xi'an

This paper analyzes the influencing factors of water demand in Xi'an,selects six variables,collects the data of total water consumption and six variables from 2004 to 2018 in Xi'an,and establishes a multiple linear regression model of water demand and six variables.After testing,the model has a high goodness of fit and overall significance,but the linear significance of coefficients with individual variables is low.Therefore,a multiple linear stepwise regression model is established by SPSS.After testing,the model can be used to simulate and predict the water consumption of Xi'an city.Finally,the multi linear step-by-step model is used to predict the total water consumption of Xi'an in 2019-2022.