k-Nearest Neighbor test
|
|
|
|
Indications/Recommendations for use:
When space-time interaction is present nearby cases will occur at about the same time, (space
nearest neighbors tend to be time nearest neighbors), and the test statistic will be large (more).
|
|
Description:
A k-nearest neighbor method used to detect space-time interaction.
|
|
Test statistic:
The number of pairs of cases that are k-nearest neighbors in both
space and time.
|
|
|
Null Hypothesis: Whether cases are nearest neighbors in space is
independent of whether they are nearest neighbors in time.
|
|
Alternative Hypothesis:
Nearest neighbors in space tend to be nearest neighbors in time.
|
|
GeoMed Inputs:
Space and time distances between pairs of cases.
|
GeoMed Outputs:
Results table includes:
- k
, the number of nearest neighbors being considered
- Jk
, the number of space-time k nearest neighbors
- significance of Jk
- DJk, the number of space-time nearest neighbors added when k increased from k-1
- significance of DJk
- p-values for the probability, under H0, of observing a statistic as large or larger than that observed and combined for the k tests in 3 ways:
- Bonferroni
- Simes
- Statistical Distance
- A linkage map
- case locations are mapped
- k
-order linkages displayed as selected by user
|
|
Example Analysis
|
Reference:
|
