1

I was working on a GWR Problem in ArcGIS to figure out the potential real estate values of Condos in a city. I thought I was following everything correctly but I keep receiving the same error over and over again.

I have attached a link to the dataset below.
The dependent variable is Finalprice. The independent (explanatory) variables are all the variables that are KM_* (distances to desirable features) and the DIST_TRECC (km to city center).

It keeps throwing the error that explains that there is a major model problem with multicollinearity. I also did the OLS regression analysis and a link to those results is below as well. Everything I read said that the numbers from the OLS analysis show that the GWR should be working. I am at a loss and not sure what to do next.

LINKS
DATASET: https://db.tt/9EUXfJy4
OLS Results: https://db.tt/LwS8pSIm

1

you are using two distance fields which can be very similar. for instance if a condo is downtown and is only .2 km from desirable features, at the same time this condo can be .2 miles from the city center, that's where your multicollineraity comes into play, use other features for hedonic modelling, ie sq ft, lot size, years on market, etc.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.