I am trying to find out whether there is any difference in equations but I am not capable to find any information about Geoda equations and to try to compare them with ArcGis equations.
Hi Vladimir,
Thats a good question. I'm not certain but I think that the results should be the same since the derivation of these statistics are well established. What could be the difference, if any, I suspect is the precision of the data types used to generate results. I'd be interested in hearing more from the community on this question.
Create a small test problem to which you know the answers and run in both software systems.
First I agree with John. It is a matter of implementation. Most software vendors hide such information; however the results are 'supposed' to be largely consistent.
Thanks to everybody for answers!
I would just want to inform you that results are not the same.
I have taken the advice from John and I have run it on the small scale area trying to test the result and I gained different result.
GeoDa and ArcGIS are producing similar but not completely the same results.
It is also maybe because of the nature of the software.
Maybe the two software packages handle feature boundaries in a slightly different way. This has happened before with the ArcGIS K-Function which produced different results for "highly" irregular boundaries.
I also find they are different. When using k-nearest neighbors spatial matrix, the difference will increase as the 'k' increasing. I guess this is because of projected coordinate system. However, which one is correct?
- Badges
- Science topic
- Similar topics
- Geoscience
- Geoinformatics