Martingale methods for the analysis of epidemic data

Area of Science:

• Epidemiology
• Biostatistics
• Mathematical Biology

Background:

• Statistical inference for epidemic models is crucial for understanding disease dynamics.
• Traditional statistical methods can be complex or unavailable for certain epidemic scenarios.
• Martingale methods offer a theoretical framework for statistical inference.

Purpose of the Study:

• To provide a tutorial introduction to martingale methods in statistical inference.
• To demonstrate the application of martingale methods for estimating parameters in epidemic models.
• To critically review existing applications of martingale methods in infectious disease data analysis.

Main Methods:

• Introduction to martingale theory using survival time data as a familiar context.
• Derivation of an inference method for the infection potential in a simple epidemic model.
• Critical review of published applications of martingale methods in infectious disease studies.

Main Results:

• Martingale methods simplify statistical inference, yielding well-known estimates and standard errors in basic settings.
• A novel, computationally simple inference method for infection potential in epidemic models was derived.
• The derived method is applicable under more general assumptions and in situations where standard methods fail.

Conclusions:

• Martingale methods provide a valuable and often simpler alternative for statistical inference in epidemic modeling.
• The approach is particularly useful when standard inference methods are cumbersome or unavailable.
• Further research is needed to confirm the high efficiency of martingale methods in various epidemic applications.