A statistical framework for measuring the temporal stability of human mobility patterns

Area of Science:

• Human Mobility
• Geospatial Data Analysis
• Statistical Science

Background:

• Human mobility studies increasingly utilize Global Positioning System (GPS) location data.
• Current statistical literature lacks guidance on the minimum required duration for GPS monitoring.
• Assessing the temporal stability of human movement patterns is essential for reliable study outcomes.

Purpose of the Study:

• To develop a theoretical framework for evaluating the temporal stability of human mobility using GPS data.
• To establish methods for analyzing human spatiotemporal trajectories and activity patterns.
• To provide evidence-based recommendations for GPS study design.

Main Methods:

• Defined measures for temporal dynamics of human spatiotemporal trajectories.
• Utilized average velocity process and activity distributions within spatial observation windows.
• Analyzed GPS location data from 185 individuals over 18 months.

Main Results:

• Empirical results indicate that current recommended GPS monitoring durations are insufficient.
• Longer monitoring periods are necessary for robust assessment of human mobility temporal stability.
• Significant differences in mobility patterns across demographic groups were observed.

Conclusions:

• GPS monitoring duration in human mobility studies should be extended beyond current suggestions.
• Future GPS study designs must incorporate considerations for demographic variations.
• The proposed framework aids in optimizing data collection for human mobility research.