Application of ARIMA and LSTM in Relative Humidity Prediction

Relative humidity is one of the important indicators of precipitation forecasting. Therefore, the prediction of relative humidity plays an important role in improving the accuracy of weather forecasting. This paper used time series and deep neural network to take meteorological data of a county in Gansu as an example, and established the ARIMA and LSTM models respectively to predict the relative humidity. By comparing the two prediction results, it was found that ARIMA had a better prediction effect on relative humidity, which could provide a favorable reference for the actual forecasting work.