Accurate short-term weather forecasting is critical for agriculture, disaster preparedness, andresource management, particularly in developing regions where computationally intensivemodels may not be feasible. This study evaluates the performance of traditional time seriessmoothing techniques for forecasting temperature, humidity, and pressure in Nsukka, Nigeria.Daily observations covering a 60-day period (October–December 2023) were obtained froman open-source mobile application linked to authoritative meteorological agencies. Methodsapplied include simple averaging, moving averages, weighted moving averages, simpleexponential smoothing (SES), and double exponential smoothing (Holt’s method). Modelperformance was assessed using error metrics including mean absolute error (MAE), meansquared error (MSE), root mean squared error (RMSE), mean absolute deviation (MAD), andmean absolute percentage error (MAPE). Results show that SES consistently outperformedother techniques for temperature and humidity forecasting, while Holt’s methoddemonstrated advantages under all conditions with pronounced trends. Pressure forecastsexhibited greater variability, suggesting sensitivity to external atmospheric influences. Thefindings highlight the practicality of smoothing techniques for short-term forecasting inresource-limited contexts, offering interpretable and computationally efficient alternatives toadvanced machine learning models. This work underscores the operational relevance ofexponential smoothing methods for meteorological services in sub-Saharan Africa.
Ugbor D. O (Mon,) studied this question.