I am running a GWR with a dependent variable that ranks from 1-5. I receive error: 110242: There is not enough variation in the dependent variable for at least one logical neighborhood. I have 5 explanatory variables with numerical ranges from 1 to several thousand. I have tried to change the Neighborhood Selection Methods and changed the ranking numbers to 100, 200, 300, 400 and 500. Neither has worked. How can I address this lack of variability since my dependent variable is a ranking so some numbers reappear literally thousands of times?