摘要
随着我国消费结构的升级与人们消费观念的变化,二手车市场交易量不断增加,但市场尚缺乏统一且透明的定价标准。因此,基于二手车平台的交易数据,构建基于PCA-DNN和LightGBM的价格预测模型,对其最终的成交价进行预测。实验结果表明,LightGBM的R2为0.8877,PCA-DNN则高达0.9479,说明模型达到了较高精度,优于LightGBM模型,具有较高的实用价值,可为解决实际中的量化问题提供参考。
With the upgrading of China’s consumption structure and the change of people’s consumption concept,the transaction volume of the used car market is increasing.However,some problems have also been exposed,and there is still a lack of a unified and transparent pricing standard in the market.Therefore,based on practical problems,this paper explores the valuation of used cars.Based on the transaction data from the used car platform,a price prediction model based on PCA-DNN and LightGBM is constructed to predict the final transaction price.The experimental results show that R~2 of LightGBM is 0.8877,and the PCA-DNN is as high as 0.9479,which shows the model has achieved high accuracy,and is better than the LightGBM,and has high practical value,which can provide a reference for solving quantitative problems in practice.
作者
杨致远
YANG Zhiyuan(School of Science,Tianjin University of Commerce,Tianjin 300134,China)
出处
《信息与电脑》
2022年第21期73-75,共3页
Information & Computer
