Forecasting New Tuberculosis Cases in Malaysia: A Time-Series Study Using the Autoregressive Integrated Moving...

Area of Science:

• Epidemiology
• Biostatistics
• Public Health Surveillance

Background:

• The Autoregressive Integrated Moving Average (ARIMA) model is a valuable tool for infectious disease prediction and public health surveillance.
• Accurate forecasting of infectious diseases like tuberculosis (TB) is crucial for effective public health interventions.

Purpose of the Study:

• To develop and validate an ARIMA model for predicting new tuberculosis (TB) cases in Malaysia.
• To forecast monthly new TB cases for the year 2019 based on historical data.

Main Methods:

• Utilized time-series data of new TB cases in Malaysia from January 2013 to December 2018.
• Applied the Box-Jenkins ARIMA methodology to identify the optimal model.
• Assessed model efficacy using Mean Absolute Percentage Error (MAPE) and stationary R-squared, validated with the Ljung-Box test.

Main Results:

• The ARIMA (2,1,1)(0,1,0)₁₂ model was identified as the most suitable, achieving a low MAPE of 6.762.
• Analysis revealed a clear seasonal pattern in new TB cases, with peaks in March and December.
• The model explained 55.8% of the variance in TB cases, with a non-significant Ljung-Box test (p=0.356).

Conclusions:

• The ARIMA model offers a simple, precise, and cost-effective forecasting tool for monitoring the TB epidemic in Malaysia.
• The model provides a valuable six-month advance warning for public health officials to manage seasonal TB trends.
• This approach enhances early warning surveillance systems for infectious diseases with predictable seasonal patterns.