Geographically Weighted Regression

Geographically Weighted Regression (GWR) fits a local regression at every location in your study area, so coefficients vary across space instead of collapsing into a single global fit. Use it when you suspect the relationship between your variables differs from one part of your study area to another — for example, after an aspatial Regression leaves significant spatial autocorrelation in its residuals (see "What to Check" on the Regression page).

GWR works on areal (polygon) data.

Coefficient Surfaces

For each explanatory variable, GWR produces a coefficient surface — a map of that variable's local effect, varying by location — alongside:

  • Local standard errors
  • Local t statistics
  • Effective degrees of freedom
  • A significance mask, showing where a coefficient's local effect is statistically significant

Process Steps

  1. Click on the "Analyze" button from the side bar menu, or search for "Geographically Weighted Regression" from the Guided Analyze wizard's Analysis step.
  2. Choose Automatic, Guided, or Full control guidance. See Analysis Introduction for more on the Guided Analyze wizard.
  3. Choose the dependent variable to explain and the explanatory variables that explain it, the same way as with Regression.
  4. Select "Run" to fit the model.
  5. Results, including a coefficient surface per explanatory variable, are saved in the Data panel to the right.