Kriging interpolation with duplicate inputs


Chris2
02-25-2008, 05:15 AM
I'm using the Krig algorithm from the 3rd edition. It works well, except when I have multiple input data points at the same location. The data points at any particular location have the exact same value (they are duplicates), so I wouldn't have thought they would pose a problem, but the interpolation generates very large excursions in the output.

Is this expected behavior, or a bug? I can, of course, sort the input data and remove duplicates -- is there an easier or better solution?

Thanks.