Forecasting Extreme High Temperatures Based on an ARIMA-LSTM Hybrid Model
Hao Chen, Zhicheng Su, Zhikai Xing
Zhengzhou University
阅读操作
确认中在文库中上传 PDF 后可生成中文音频讲解。
摘要与影响
This paper employs the ARIMA model and LSTM model to forecast and compare the occurrence of extremely hot weather in the region, respectively. Addressing the issue of low precision and inaccurate forecasts in existing models for predicting extremely hot weather, the ARIMA model's forecasts are refined using the LSTM model's error predictions, resulting in the proposed ARIMA-LSTM hybrid model. The findings indicate a significant improvement in the ARIMA model's forecast accuracy across the metrics of RMSE, MAE, and goodness-of-fit$\mathbf{R}^{\mathbf{2}}$. The ARIMA-LSTM model's predictions more accurately reflect actual conditions compared to the individual ARIMA and LSTM models. Consequently, developing an appropriate ARIMA-LSTM model for forecasting extreme high temperatures, tailored to the region's specific conditions, could provide a theoretical basis for preventing hot weather and devising mitigation strategies.
逐年被引趋势
关键指标
同类平均 = 1
同领域 · 同年份 · 同类型
Google Scholar 与 OpenAlex 的被引统计范围不同,数值存在差异属正常。
AI 辅助阅读
依据:摘要
可就本文提问;依据不足时会说明。
学术脉络
学科主题
物理Hydrological Forecasting Using AI
Stock Market Forecasting Methods · Forecasting Techniques and Applications