合肥工业大学校徽 合肥工业大学学报自科版

导航菜单

基于函数型期望分位数回归森林模型的 AQI 预测

Prediction of air quality index based on functional expectile regression forest model

期刊信息

合肥工业大学(自然科学版),2025年6月,第48卷第6期:823-827

DOI: 10.3969/j.issn.1003-5060.2025.06.017

作者信息

陈慧琪,凌能祥

(合肥工业大学数学学院,安徽合肥230601)

摘要和关键词

摘要: 文章将函数型数据分析和期望分位数回归森林(expectile regression forest, ERF)模型相结合,分析了合肥市2015—2022年空气质量,并利用函数型ERF模型对空气质量指数(air quality index, AQI)进行预测。研究结果表明,大部分真实值均落在预测区间中,期望分位数回归森林模型表现出较好的预测结果,体现出函数型数据与随机森林模型相结合的优势。

关键词: 函数型数据;期望分位数回归;随机森林;非参数回归;空气质量指数(AQI)

Authors

CHEN Huiqi, LING Nengxiang

(School of Mathematics, Hefei University of Technology, Hefei 230601, China)

Abstract and Keywords

Abstract: In this paper, the air quality in Hefei City from 2015 to 2022 was analyzed by combining functional data analysis (FDA) and expectile regression forest (ERF) model, and the air quality index (AQI) was predicted based on functional ERF model. It is found that most of the actual values fall within the prediction interval, indicating that the AQI of Hefei City is well predicted by ERF model, which exhibits the advantages of FDA with random forest model.

Keywords: functional data; expectile regression; random forest; non-parametric regression; air quality index(AQI)

基金信息

国家自然科学基金资助项目(72071068)

个人中心