On the statistical description of large populations

Area of Science:

• Mathematical modeling
• Statistical physics
• Population dynamics

Background:

• Existing population dynamics models often lack clear interconnections between heuristic and computer-based methods.
• The effectiveness of computational approaches is limited by vague interconnections in current population dynamics research.
• Typically, population dynamics are described using only low-order correlations.

Purpose of the Study:

• To propose and justify the use of higher-order correlations for population dynamics.
• To introduce probability measures as micro-states for analyzing correlation functions.
• To determine conditions under which low-order correlations are insufficient for population modeling.

Main Methods:

• Analysis of an individual-based model for an infinite population of interacting entities.
• Explicit use of probability measures as micro-states.
• Derivation and analysis of evolution equations for correlation functions.
• Numerical solution and analysis of the derived kinetic equation.

Main Results:

• Demonstrates the essential role of higher-order correlations in populations with strong local interactions.
• Identifies sub-Poissonian states where particle distribution can be described by density.
• Shows that the analyzed model preserves sub-Poissonian states.
• Obtained and analyzed a kinetic equation for the population dynamics.

Conclusions:

• Higher-order correlations are crucial for accurately modeling certain population dynamics.
• Probability measures as micro-states offer a framework for analyzing correlation functions.
• The developed kinetic equation provides insights into the evolution of sub-Poissonian states in interacting populations.