Time Series Analysis of Mean Temperature using SARIMA: An Example from Davao Oriental, Philippines
Keywords:forecasting, mean temperature, climate change, time series, Davao Oriental
In many practical disciplines, time series analysis, and forecasting—a technique that predicts future values by analyzing past values—play a substantial role. In this paper, the researchers analyze the monthly mean temperature in Davao Oriental from 2010 to 2022 using the \SARIMA (Seasonal Auto-Regressive Integrated Moving Average) technique. Data from January 2010 to May 2020 were used as the training data set, while data from June 2020 to December 2022 were used as the testing data set. The presentation includes a thorough overview of model selection and forecasting precision. The findings demonstrate that the suggested research strategy achieves good forecasting accuracy. The analysis reveals that the best model which was satisfactory to describe was SARIMA (0,1,3) (2,0,0) , and in the month of May 2023, the temperature will be 28.27964 degrees Celsius. In subsequent work, the researchers hope to expand the number of possible grid search parameter combinations. This method may lead us to models with improved predictive ability. The length of the training set may also affect forecasting accuracy, in addition to the SARIMA model's parameters. A follow-up study is needed to investigate both hypotheses.
How to Cite
Copyright (c) 2023 Larry M. Ichon Jr., Jerd M. Dela Gente
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
DRJ is an open-access journal and the article's license is CC-BY-NC. This license allows others to distribute, remix, tweak, and build on the author's work, as long as they give credit to the original work. Authors retain the copyright and grant the journal/publisher non-exclusive publishing rights with the work simultaneously licensed under a https://creativecommons.org/licenses/by-nc/4.0/.