BayesianFitForecast

BayesianFitForecast streamlines Bayesian parameter estimation and forecasting with ordinary differential equations for dynamical systems in health informatics and public health decision-making.

Key Features:

• Automatic Stan File Generation: The toolbox automatically generates Stan files, enabling users to configure models, define priors, and analyze results without manual coding.
• Model Performance Evaluation: Provides convergence diagnostics, posterior distributions, credible intervals, and performance metrics for assessing model results.
• Multiple Observation Error Structures: Supports parameter estimation and forecasting under Poisson and negative binomial observation error structures.

Scientific Applications:

• Epidemic Modeling: Demonstrated on historical epidemic datasets including the 1918 influenza pandemic in San Francisco and the 1896-1897 Bombay plague.
• Healthcare Forecasting: Enables time-series modeling and forecasting applications in healthcare and epidemiological contexts.
• Simulated Data Analysis: Validated using simulated data to demonstrate robust parameter estimation and forecasting performance.

Methodology:

Bayesian inference with ordinary differential equations implemented through automatic Stan file generation; users configure models, define priors, and analyze results through the R toolbox interface.