Related Experiment Videos
Intransitivity, the spatial interaction model, and U.S. migration streams
Environment & Planning A
|January 1, 1980
Summary
Spatial interaction models predict transitive flows, but US migration data shows intransitivities. This study explores conditions causing these deviations in migration patterns.
Area of Science:
- Spatial analysis
- Human geography
- Econometrics
Background:
- Spatial interaction models commonly predict transitive flows.
- US interstate migration data exhibits significant intransitivities.
- Understanding these deviations is crucial for accurate migration modeling.
Purpose of the Study:
- Investigate conditions leading to intransitive flows in spatial interaction models.
- Analyze factors causing observed intransitivities in US migration data.
- Calibrate and compare spatial interaction models with empirical migration data.
Main Methods:
- Calibration of a singly constrained gravity model.
- Distortion of flow tables by sampling error.
- Aggregation over strata and introduction of independent error terms.
Main Results:
- Demonstration of how sampling error and aggregation can induce intransitivities.
- Calibration results for US interstate migration flows (1935-1970).
- Comparison of calibrated model results with existing literature.
Conclusions:
- Sampling error and aggregation are significant contributors to observed intransitivities.
- Spatial interaction models require adjustments to accurately reflect real-world migration patterns.
- Further research needed to refine models for complex migration dynamics.