Sarima Modeling of Monthly Temperature in the Northern part of Ghana

Abstract

The Sarima model is used in this study to forecast the monthly temperature in Ghana's northern region. The researchers used temperature data from January 1990 to December 2020. The temperature data was found to be stationary using the Augmented Dickey Fuller (ADF) test. The ACF and PACF plots proposed six SARIMA models: SARIMA (1,0,0) (1,0,0) (12), SARIMA (2,0,0) (1,0,0) (12), SARIMA (1,0,1) (1,0,0) (12), SARIMA (0,0,1) (1,0,0) (12), SARIMA (0,0,1) (0,0,1) (12), SARIMA (0,0,1) (0,0,1) (12). The best model was chosen based on the lowest Akaike Information Criteria (AICs) and Bayesian Information Criteria (BIC) values. The Ljung-Box data, among others, were used to determine the model's quality. All diagnostic tests are passed by the SARIMA (1,0,0) (1,0,0) (12) model. As a result, the SARIMA (1,0,0) (1,0,0) (12) is the best-fitting model for predicting monthly temperatures in Ghana's northern region.

A Comparative Study of Arima and Sarima Models to Forecast Lockdowns Due to Covid-19

“…The study on the SARIMA model has shown that this model has various uses in forecasting data related to -Weather and Temperature forecasting [33], Tourism [34], fire frequency or accidents forecasting [35], the financial sector [36], and healthcare industry [37]. Talking about the healthcare industry, the current paper deals with the analysis and forecasting of region-wise smart lockdowns.…”

“…Some of the commonly used examples in the Machine Learning domain have been hereby stated with a small summary of the objective and the results produced by SARIMA Models as various different applications. Paper [33] describes a SARIMA Model capable of forecasting monthly temperature values in Ghana's northern area. The SARIMA(1,0,0)(1,0,0)( 12) is the best-fitting model, according to the results.…”

“…There are various machine learning algorithms that are used for forecasting. And here is a study [33] that predicts monthly precipitation data in Bandung. A comparison among extreme gradient enhancement models, SARIMA, neural networks (NN), and exponential smoothing (ETS),.…”