The following example calculates the local Moran’s I statistic for each row in a table from a specific column and uses the spatial weights already saved in a datastore. It uses the median_income column and the spatial weights from the spatial datastore with the object name la_bg_knn4, corresponding to the spatial weights calculated with the K-nearest neighbors method with K=4.

select *
    from table( 
        pyqEval(
            '{  
                "oml_connect": true, 
                "table": "oml_user.la_block_groups", 
                "key_column": "geoid",
                "column": "median_income",
                "weights": {"ds_name":"spatial", "obj_name": "la_bg_knn4"}
            }',
            '{ "geoid": "VARCHAR2(50)", "I": "NUMBER", "p_value": "NUMBER",  "z_value": "NUMBER", "quadrant": "NUMBER" }',
            'compute_local_spatial_autocorrelation'
        )
    );

For each row in the table, the result contains the following:

• The local Moran’s I statistic.
• The p-value.
• The z-value.
• The belonging quadrant.
  1. A high value surrounded by high values.
  2. A low value around high values.
  3. A low value surrounded by low values.
  4. A high value around high values.