[SOLVED] Auto-fitted Variogram Model parameters

[Jeffrey Yarus]

If I am using autofitting and would like to understand the final parameters, particularly for the Stable Variogram:
#### Model characteristics## =====================
## Space dimension= 2
## Number of variable(s)= 1
## Number of basic structure(s) = 1
## Number of drift function(s) = 129 - HOW CAN I SEE THE DETAIL IF I WANTED TO?
## Number of drift equation(s) = 1
#### Covariance Part
## ---------------
## - Stable
## Range= 22108.486 -----WHAT IS THE DIFFERENCE BETWEEN THIS AND THE THEO. RANGE?
## Theo. Range = 12225.614 ---- WHAT IS THE DIFFERENCE BETWEEN THIS AND THE RANGE?
## Parameter =1.854 ----THIS MUST BE THE POWER TERM?
## Sill= 27047.156
## Total Sill = 27047.156
#### Drift Part## ----------
## Universality Condition

[Didier Renard]

The first question is how did you manage to obtain the number of drift functions equal to 129. This is completely unrealistic and may result from a previous bug.

The second question concerns the difference between theoretical and practical ranges. Just to make it simple:
- the theoretical range is the value which appears in the formula:

For example in the Exponential Basic structure, C(h) = C exp(- h/a) where a is the theoretical range

- the practical range is the distance at which the variogram reaches 95% of the total variance (or when the covariance reaches 5% of the variance).
Obviously the difference between the two ranges only holds in the case of functions which reach their sill asymptotically. For the other basic structures, these two values are equal.

Finally note that the relationship between theoretical and practical ranges may depend upon the value of the third coefficient.