基于BP神经网络的居民消费水平的预测模型——以安徽省为例  

Residents′consumption level prediction model based on BP neural network:taking Anhui Province as an example

作　　者：路思恒 尹红[1] LU Siheng;YIN Hong(Faculty of Mechanical and Electrical Engineering,Kunming University of Science and Technology,Kunming 650500,China)

机构地区：[1]昆明理工大学机电工程学院,云南昆明650500

出　　处：《现代电子技术》2022年第21期83-87,共5页Modern Electronics Technique

基　　金：云南智能化自动化产业发展研究(云府发研【2017】32号—YNDR2017G1C06)。

摘　　要：研究并分析居民消费水平和影响因素对我国经济的发展格外重要。基于BP神经网络建立居民消费水平预测模型,经过研究分析大量的相关文献,最终选择9个指标作为输入变量,居民消费水平作为输出变量;选择安徽省1990—2018年的统计数据作为搭建预测模型的训练集和测试集。首先搭建神经网络,输入训练集,对神经网络进行不断的学习,运用梯度下降法,使损失函数达到最小的情况下,参数即为最优,也就确定了最终的神经网络预测模型;其次将归一化后的测试集输入到训练完毕的神经网络预测模型,得出预测值,并与真实值进行对比,得出误差率;最终该神经网络预测模型在训练集和测试集上的误差率均小于10%,达到了预期的结果,说明具有很好的可行性。It is particularly important to study and analyze residents′consumption level and influencing factors for China′s economic development.In this paper,a prediction model of residents′consumption level is established based on BP neural network.After studying and analyzing a large number of relevant literature,9 indexes are finally selected as input variables and residents′consumption level as output variables.The statistical data of Anhui Province from 1990 to 2018 are selected as the training set and test set for building the prediction model.The neural network is built,the training set is input,and the neural network is continuously studied.The gradient descent method is used to minimize the loss function.When the loss function reaches the minimum,the parameters are the optimal,which determines the final neural network prediction model.The normalized test set is input into the trained neural network prediction model to obtain the predicted value.And then,the predicted value is contrasted with the real value to obtain the error rate.The error rate of the neural network prediction model is less than 10%in the training set and test set respectively,which achieves the expected results.Therefore,the model is of feasibility.

关 键 词：预测模型 BP神经网络 居民消费水平 SIGMOID函数 梯度下降法 损失函数 数据归一化 

分 类 号：TN711-34[电子电信—电路与系统]