Prediction of Lung Cancer Patients' Numbers at Sabratha Oncology Hospital Using Exponential Smoothing Models

Abstract

Cancer, particularly lung cancer, is one of the major global health challenges. This study aims to analyze the historical trends of lung cancer cases in Libya from 2005 to 2023 using statistical forecasting models. Three statistical models were applied to predict future lung cancer cases: the Simple Exponential Smoothing model, Holt’s Linear Trend model, and the Damped Trend model. The results of the analysis showed that the Simple Exponential Smoothing model was the least accurate, while the Holt model provided better results but was not the best in terms of prediction accuracy. In contrast, the Damped Trend model exhibited the lowest RMSE value (78.20) and demonstrated high accuracy in forecasting future data, making it the most reliable model. The significance of the model was also tested, and the results revealed strong statistical significance for the parameters, further validating its reliability for forecasting. Based on these findings, it is recommended to use the Damped Trend model for guiding future health policies and improving strategies for the prevention and treatment of lung cancer.