Forecasting New Tuberculosis Cases in Malaysia: A Time-Series Study Using the Autoregressive Integrated Moving Average (ARIMA) Model

Background The application of the Box-Jenkins autoregressive integrated moving average (ARIMA) model has been widely employed in predicting cases infectious diseases. It shown a positive impact on public health early warning surveillance due to its capability producing reliable forecasting values. This study aimed develop prediction for new tuberculosis (TB) using time-series data from January 2013 December 2018 Malaysia and forecast monthly TB 2019. Materials methods ARIMA was executed gathered between Malaysia. Subsequently, well-fitted make projections year To assess efficacy model, two key metrics were utilized: mean absolute percentage error (MAPE) stationary R-squared. Furthermore, sufficiency validated via Ljung-Box test. Results results this revealed that (2,1,1)(0,1,0)12 proved be most suitable choice, exhibiting lowest MAPE value 6.762. showed clear seasonality with peaks occurring March December. proportion variance explained by 55.8% p-value (Ljung-Box test) 0.356. Conclusions developed simple, precise, low-cost provides six months advance monitoring epidemic Malaysia, which exhibits seasonal pattern.