I know that there are several types of kernel density functions, such as uniform, cosine, Gaussian, etc. I was just wondering which one the "Kernel Density" functionality in ArcMap 10's Spatial Analyst uses?
-
I reverse-engineered this many years ago (in Spatial Analyst 1.0); a detailed, illustrated explanation is at quantdec.com/SYSEN597/GTKAV/section9/density.htm. Presumably it still applies. – whuber Aug 28 '12 at 16:06
Silverman quadratic "The kernel function is based on the quadratic kernel function described in Silverman (1986, p. 76, equation 4.5)."
The reference is to this book:
http://books.google.com/books?id=e-xsrjsL7WkC&printsec=frontcover#v=onepage&q&f=false
on page 76, equation 4.5
It seems to me that it uses quartic kernel function as below:
Where f(j) is the density at the center of the cell, h is Search Radius, dij is the distance between the incident point and center of the cell and n is the number of incidents within the search radius.
