Skip to main content
Cross Validated

Return to Question

edited tags
Link
whuber ♦
whuber ♦
  • 343.7k
  • 66
  • 823
  • 1.4k
deleted 1 character in body
Source Link
Aaron de Windt
Aaron de Windt
  • 221
  • 1
  • 4

I have to calculate the 2d kernel density estimate (kde) from a list of latitude and longitude coordinates. But one degree in latitude is not the same distance as one degree in longitude, this means that the individual kernels would be oval, specially the further the point is from the equator.

In my case the points are all close enough to each other that transforming them to flat earth, should not cause muchmany problems. However I'm still curious on how this should be properly handled in case this wasn't true.

I have to calculate the 2d kernel density estimate (kde) from a list of latitude and longitude coordinates. But one degree in latitude is not the same distance as one degree in longitude, this means that the individual kernels would be oval, specially the further the point is from the equator.

In my case the points are all close enough to each other that transforming them to flat earth, should not cause much problems. However I'm still curious on how this should be properly handled in case this wasn't true.

I have to calculate the 2d kernel density estimate (kde) from a list of latitude and longitude coordinates. But one degree in latitude is not the same distance as one degree in longitude, this means that the individual kernels would be oval, specially the further the point is from the equator.

In my case the points are all close enough to each other that transforming them to flat earth should not cause many problems. However I'm still curious on how this should be properly handled in case this wasn't true.

Source Link
Aaron de Windt
Aaron de Windt
  • 221
  • 1
  • 4

What is the proper way of calculating the kernel density estimate from geographical coordinates?

I have to calculate the 2d kernel density estimate (kde) from a list of latitude and longitude coordinates. But one degree in latitude is not the same distance as one degree in longitude, this means that the individual kernels would be oval, specially the further the point is from the equator.

In my case the points are all close enough to each other that transforming them to flat earth, should not cause much problems. However I'm still curious on how this should be properly handled in case this wasn't true.