StatModPredict

StatModPredict enables fitting and forecasting of time series data using auto-regressive integrated moving average (ARIMA), generalized linear models (GLM), generalized additive models (GAM), and Meta's Prophet model for real-time and retrospective forecasting analysis.

Key Features:

• Model Fitting: Supports fitting of ARIMA, GLM, GAM, and Prophet models to incident time series data with customizable parameters.
• Forecasting: Generates forecasts with customizable forecasting options and parameters.
• Model Evaluation: Provides evaluation metrics for model fit and forecast performance.
• Visualization: Produces multiple editable figures for time series data, forecasts, and model fit and forecast metrics.
• External Forecast Comparison: Allows upload of externally produced forecasts for performance evaluation alongside built-in models.
• Retrospective Analysis: Supports retrospective model analysis in addition to real-time forecasting.

Scientific Applications:

• Epidemic Forecasting: Forecasts epidemic trajectories using publicly available health data such as annual HIV case data in the US.
• Public Health Surveillance: Supports real-time and retrospective forecasting efforts for public health monitoring and policy planning.
• Cross-Disciplinary Time Series Analysis: Applicable to any discipline employing time series data due to flexible input data structure requirements.

Methodology:

Users load an incident time series data set into the interface, customize model parameters and forecasting options, and obtain output including fitted models, forecasts, evaluation metrics, and visualizations.