An empirical example on selecting a search radius in the ArcGIS kernel density tool is attached for your consideration.   

Article Where the bears roam in Majella National Park, Italy

Thank you so much sir

Hello Arun, I don't like default outputs Kernel estimator. They are beautiful but in general not accurate. I work with leishmaniasis diseases and I use specific radius thinking in vector disease (200 meters). In other cases, I try a radius based on the phenomenon and compare it... if the results is not too much different from default radios, I use it. 

Thanks Patricia,  I appreciate that you have answered directly. Thanks once again

Thanks Joaquín ! Thats a nice paper