基于ARIMA和GM模型的日最高温预测

The Prediction of Daily Maximum Temperature Based on ARIMA and GM Models

摘要: 基于差分整合移动平均自回归模型(Autoregressive Integrated Moving Average Model, ARIMA)、灰色预测模型(Gray Model, GM)、ARIMA和GM组合模型(ARIMA-GM)来预测夏季日最高温度。本文收集了某市2019~2021年6~8月日最高温数据作为实验数据。实验结果表明，GM、ARIMA-GM组合模型具有更好的预测结果。与GM模型相比，ARIMA-GM组合模型的预测结果更加稳定。

Abstract: Based on the Autoregressive Integrated Moving Average model (ARIMA), Gray Model (GM), ARIMA and GM combined models (ARIMA-GM), the daily maximum temperature in summer is predicted. This paper collects the maximum temperature data in a city from June to August in 2019~2021, and used it as experimental data. The experimental results show that the GM model and ARIMA-GM model have better prediction results. Compared with the GM model, the prediction results of the ARIMA-GM model are more stable.