Time Series Analysis and Prediction from Temperature Anomaly

This repository contains the code and documentation for making the predictive machine learning study, as focuses on forecasting global temperature anomalies using ARIMA and SARIMA models.

Global temperature anomalies refer to deviations in temperature from historical averages. Understanding these anomalies is essential for analyzing climate variability and making informed predictions about future climate trends.

This project uses time series data from NOAA, spanning from the year 2000 to 2024 (monthly), to build ARIMA and SARIMA models to forecast temperature anomalies. The dataset consists of three categories:

• Global Land and Ocean
• Global Land Only
• Global Ocean Only

By employing ARIMA and SARIMA models, the goal of this project is to predict future temperature trends, comparing the performance and accuracy of both models.

For more information, please visit at this link