Font Size: a A A

Research On Second-hand Car Valuation Model Based On GA-MIV-BP Algorithm

Posted on:2021-03-24Degree:MasterType:Thesis
Country:ChinaCandidate:S R YangFull Text:PDF
GTID:2392330602480351Subject:Asset appraisal
Abstract/Summary:PDF Full Text Request
With the increasing volume of second-hand cars,it is urgent to establish a fair and orderly second-hand car evaluation market.Standard evaluation platform and reasonable and effective evaluation methods are conducive to the healthy development of second-hand car evaluation industry.The research on second-hand car evaluation in China is still in the exploratory stage,and the traditional evaluation methods,with complicated evaluation steps,strong subjectivity and insufficient accuracy,cannot meet the market demand.The characteristics of second-hand car evaluation determine that only "one person and one car" can be used in practical work,instead of batch evaluation,which greatly reduces the evaluation efficiency.Therefore,it is of great significance to find an efficient and accurate evaluation method to promote the development of second-hand car evaluation industry.This paper firstly analyzes the background of second-hand car evaluation market at home and abroad,studies the three traditional evaluation methods and their advantages and disadvantages,and determines the research route of the whole paper.Secondly,it studies many factors that influence the value of second-hand cars and selects 12 of them as model indicators.In view of the shortcomings of the existing methods,this paper puts forward the use of artificial neural network self-learning ability and strong nonlinear mapping ability to establish a model,combined with the evaluation of second-hand cars.At the same time,genetic algorithm and variable screening are used to optimize and improve the model to ensure the accuracy of the evaluation results,thus determining the research method of this paper.Finally,according to the collected Jiugongli used car trade market data in Banan distract of Chongqing city as the model sample,using MATLAB software to training and testing the model,by choosing the actual cases,the GA-MIV-BP neural network model with the current market value method and replacement cost method,this paper compares and analyzes this model was verified in the used car price evaluation of the effectiveness,feasibility and convenience.
Keywords/Search Tags:second-hand car evaluation, second-hand car price, influencing factors, Back Propagation, genetic algorithm, variable screening
PDF Full Text Request
Related items