GWR for Claims Data

11-24-2021 01:17 PM
New Contributor III


I am trying to apply GWR on patients' claims data. for every record of patients, I have demographic info and their ZIP Codes. so if I plot them, they will be mapped on each other at the centroid of ZIP Codes.

1. GWR works on my data but I wonder whether the results are reliable for points data like this?

2. I have a binary outcome and several binary independent variables. While I can choose Binary for the outcome variable in the GWR settings, can I use binary independent variables (or even categorical variables with more than 2 classes) in GWR? if yes, I wonder how I should interpret the results for independent variables' coefficients?  let's say I code males as 0 and females as 1. so the coefficients should be interpreted for females as they are coded 1? what if they are coded as males with 1 and females with 2?

Thanks for the feedback in advance.

0 Kudos
0 Replies