# What percent of non-zero weights should I aim for when selecting value of k?

I am attempting to test for spatial-autocorrelation among regression residuals via Moran's I in R using the 'spdep' package. While creating the spatial weights, changing the value of k changes the result of my Moran's I test. I have 1582 regions, and running the following code:

>knn2nb(knearneigh(spoints,k=100))


returns:

>Neighbour list object:
Number of regions: 1582