AI-GEOSTATS: fitting a variogram
- Hi list.
I have fitted the variogram of my data set but i'm very unsure with the results.
I'm using arcinfo software to make the geostatistical analysis.
The software fit automatically the variogram to the data but i change this fitting manually in the way i think.
The software fitting is based on minimizing the error and i think that this is an objective way to fit.
The user fitting is based on experience (i think) and the mine is poor, so is an subjective way to fit.
I will thank any suggestion that can help myself make best decissions about fitting my data.
Javier Lastra Fernández
Sistemas de Información Geográfica
Campus de Mieres, 33600. Mieres
Teléfono: Int + 985458114
[Non-text portions of this message have been removed]
Did you try more than one "manual" fitting? Which has the
smallest error???? The first fitting the ArcGIS is doing is very
"automatic" if i can say it - and you can do manual fittings with
better errors (i mean smaller errors). Try more than one kriging
method, change the neighbourhood and the min and max number
of known values in the neighbourhood, see if directional kriging will
not be better and so on. The idea is that you do lots of kriging
using different semi-variograms and you record their parameters
and what errors you will get. Do a cross-validation and see which
performs better. Afterwards you can compute the prediction
confidence intervals by one of the methods available - like
likelihood or bootstrapping or bayesian .... some or all of them are
implemented in R - a free environment very similar with S-PLUS.
I hope this helps,
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and "unsubscribe ai-geostats" followed by "end" on the next line in the message body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org