Usually, variable values are interpolated with IDW and Kriging methods. Is there any method to verify these methods' resulted maps? suppose that, drawing climatic maps, how can we sure that point x in real conditions have that value? Do linear regression can be useful in this field?
http://desktop.arcgis.com/en/arcmap/latest/extensions/geostatistical-analyst/performing-cross-validation-and-validation.htm
read page 11of below link for cross-validation diagram gives you an idea of how well the model predicts the values at the unknown locations.
http://help.arcgis.com/en/arcgisdesktop/10.0/pdf/geostatistical-analyst-tutorial.pdf
Dear Dr. Firoz Ahmad
Please accept my special thanks for your constructive response.
Depends on the data format you have. If it is on vector format, it is easier to export validation and predicted datasets into SPSS or R Softaware acceptable format and perform regression analysis. In case both predicted and validation are on raster format, you first need to convert them into vector format, then follow the normal procedure as stated before. This may best suit for numerical data. For categorical data, you neen to define both of the datasets on same spatial scale and use SPSS to perform regression analysis. Fundamentals of experimental design of important in this regard
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Publication