I have a scatter of points (longitude, latitude). There is a lot of spread obviously, but also a certain degree of congestion. Any idea how I can estimate something like a 95 % region where I expect the points to lie?

Just looking for reading material here. The word "kernel" is being thrown around on the internet, but I can't really fin something specific.


The points are GPS locations of home addresses that a delivery company has delivered their product to over a period of a year.

The interest lies in determining a region, say a x % region, where we expect the next delivery to lie in.

  • 3
    $\begingroup$ Could you please clarify your objectives, the context, and the nature of your data? Since you already have points, what exactly does it mean to "expect the points to lie"? For instance, suppose your points are locations of the capitals of all countries in the world. What would "where I expect the points to lie" mean in that case? $\endgroup$ – whuber Apr 27 '17 at 18:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.