A non-central beta model to forecast and evaluate pandemics time series

Area of Science:

• Epidemiology
• Biostatistics
• Time Series Analysis

Background:

• Pandemic modeling (e.g., COVID-19) presents challenges due to novelty and local-dependent spread.
• Existing models (SIR, SEIR, ARIMA, ETS) have limitations in capturing pandemic dynamics.
• Accurate forecasting and evaluation of pandemic time series are crucial for public health response.

Purpose of the Study:

• To propose a simple and effective framework for modeling, forecasting, and evaluating pandemic incidence time series.
• To utilize the non-central beta (NCB) probability density function for pandemic modeling.
• To compare the proposed NCB framework against established epidemic and time series models.

Main Methods:

• Development of a probabilistic optimization algorithm to fit the NCB model.
• Minimization of the mean square error (MSE) to select the best-fitting NCB model.
• Application of the framework to COVID-19 incidence time series data from various countries.

Main Results:

• The NCB model successfully infers key pandemic metrics, including peak date, end date, and total cases.
• The framework demonstrates utility in comparing pandemic severity across different territories.
• Case studies show the NCB framework's effectiveness compared to SIR, SEIR, ARIMA, and ETS models.

Conclusions:

• The proposed NCB framework offers a valuable tool for pandemic analysis and forecasting.
• The NCB model provides a robust alternative to traditional epidemic and time series models.
• This approach aids in understanding and managing the impact of novel infectious diseases.