Uncertainty analysis of contagion processes based on a functional approach.

Area of Science:

• Epidemiology
• Network Science
• Stochastic Processes

Background:

• Predicting the spread of diseases, products, or ideas in populations is inherently difficult.
• Limited observations of contagion processes hinder accurate forecasting of future events.
• The stochastic nature of contagion leads to unpredictable outcomes, potentially causing policy inaccuracies.

Purpose of the Study:

• To analyze the unpredictability and uncertainty in contagion processes.
• To develop a novel non-parametric measure of variance for functional data.
• To investigate the impact of network properties and process contagiousness on uncertainty.

Main Methods:

• Extensive simulation study of contagion dynamics.
• Definition of a novel non-parametric variance measure using weighted depth-based central regions.
• Application to the susceptible-infected-susceptible (SIS) epidemiological model on small-world networks.

Main Results:

• Maximum uncertainty in contagion is observed at the epidemic threshold.
• Network density and process contagiousness significantly and complementarily affect contagion uncertainty.
• The network's randomness structure has a minor impact on contagion uncertainty.

Conclusions:

• Understanding contagion uncertainty is crucial for accurate predictions and effective policymaking.
• The developed variance measure provides a novel approach to quantify uncertainty in functional contagion data.
• Network structure and transmission dynamics are key drivers of contagion unpredictability.