Skip to main content

​
g_local

Calculates the Getis-Ord Gi or Gi* statistic on the x column of the dataframe.

​
Additional Information

For more information on the Getis-Ord functions see: “Getis-Ord functions. From the 1992 paper by Getis & Ord. Getis, A., & Ord, J. K. (1992). The analysis of spatial association by use of distance statistics. Geographical Analysis, 24(3), 189-206. https://doi.org/10.1111/j.1538-4632.1992.tb00261.x

​
Parameters

​
DataFrame
required
the dataframe to perform the G statistic on
​
str
required
The column name we want to perform hotspot analysis on
​
str
required
The column name containing the neighbors array. The array should be of type Array<Struct<value: T, neighbor: U>>, where each element is a Struct with two fields:
  1. value: The weight for the neighbor (e.g., spatial weight).
  2. neighbor: A Struct containing the data of the neighboring row (schema must match the parent DataFrame, excluding the weights column itself). You can use wherobots.weighing.add_distance_band_column to generate this column in the required format.
​
int
required
Not used. Permutation tests are not supported yet. The number of permutations to use for the
​
bool
required
Specifies whether to calculate the Getis-Ord Gi* statistic.
  • true: Calculates the Gi* statistic, which includes the focal observation (the row itself) in the local sum. When star=true, the weights array must include the focal observation as one of its own neighbors.
  • false: (Default) Calculates the G-statistic, which excludes the focal observation.
​
float
required
Not used. The weight for the simulated neighbor used for records without a neighbor in perm testsPermutation testing is not yet implemented. Consequently, any parameter related to island_weight currently has no effect.

​
Returns

A DataFrame with the original columns plus the following additional columns:
​
Double
The calculated local G or G* statistic.
​
Double
The expected value of G/G* under spatial randomness.
​
Double
The variance of G/G* under spatial randomness.
​
Double
The Z-score (standard score) for the statistic. Statistical significance (Z and P columns) is calculated using Z-scores derived from the theoretical expected value and variance under spatial randomness.
​
Double
The p-value (significance) derived from the Z-score. Statistical significance (Z and P columns) is calculated using Z-scores derived from the theoretical expected value and variance under spatial randomness.

​
Usage Examples