Using SARIMA Modeling and Forecasting of Metrological Parameters: A Conceptual Framework

Abstract

Objective: The current study aims to predict the metrological factor of temperature in the region of Karachi Methodology: using the Seasonal Auto-Regressive Integrated Moving Average (SARIMA) model. Daily maximum and minimum temperature data from the region from 1st January, 2012 to 31st December, 2022 using as training data for the model. These datasets are clean and modify to obtain monthly averages of the maximum and minimum temperature data for the region. These datasets using for the model development. The two datasets pass through time series analysis separately and best fitted models are developing for both. Finding: The study shows the presence of seasonality in the temperature data alongside the presence of a growing mean in the minimum and maximum temperature dataset, signifying global warming over the past decade. Implications: These results will help in future studies regarding the variations in temperature in Karachi and in developing strategies to accommodate for such variations in the region