I am new to Cluster analysis, and I’m using a spatial descriptive statistical method called the Ripley’s (K) function for my analysis purposes. I read some really good papers which explained the applications of this function, but I’m still baffled with some questions;
(1) What is the purpose of using the border correction algorithms?
(2) What are the information we are able to extract from the spatial distribution graph plotted with "r" (distance) Vs "L(r) – r"?
Dear Dilshan
This method considers virtual concentric circles around each point and counts the number of points that fall into each circle. If the pattern follows random distribution, we can expect the number of point in each circle with radius of r would be equal and if you draw this numbers cumulatively againts r, you will have a 45 degree line in your graph. Higher degree lines show clustered pattern. So, if you have a small study area, the circles will be truncated. The edge correction methods are deal with this problem. K should be standardized based on the density of points. The formula which do this standardization yields L (r) values.
best regards,
Hormoz
- Badges
- Science topic