Analysis and Forecasting of Monthly Average Temperatures Using Timeseries Models

Accurate temperature forecasting is crucial in various domains, including meteorology, agriculture, and urban planning. This research aims to determine the most precise time series model for predicting future temperatures in prominent Indian cities such as Mumbai, Kolkata, Chennai, and Delhi. The assessment of three distinct models, namely Holt-Winters with Exponential Smoothing, SARIMAX, and Prophet, using the Root Mean Squared Error (RMSE) statistic. The primary objective is to identify the model with the lowest RMSE values, indicating superior accuracy in capturing temperature patterns. The outcomes of this study can significantly assist decision-making and planning processes in industries that heavily rely on precise temperature forecasts.