next up previous contents
Next: Example 3.1: Deposit of Up: Stochastic Hypotheses Previous: Stationarity of Order 2   Contents

Intrinsic Hypothesis

A random function $Z({/boma x})$ is said to be intrinsic if:

(i)
the mathematical expectation exists and does not depend on the support point $/boma{x}$,

/begin{displaymath}E[Z({/boma x})]=m, / / /forall {/boma x}/ / / ;/end{displaymath}

(ii)
for all vectors ${/boma h}$ the increment [ $Z({/boma
x}+{/boma
h})-Z({/boma x})$] has a finite variance which does not depend on ${/boma x}$,

/begin{displaymath}Var[Z({/boma x}+{/boma h})-Z({/boma x})]=
E[Z({/boma x}+{/boma h})-Z({/boma x})]^{2}=2/gamma({/boma h})/ / / ./end{displaymath}

Thus, second-order stationarity implies the intrinsic hypothesis, but the converse is not true: the intrinsic hypothesis can also be seen as the limitation of the second-order stationarity to the increments of the random function $Z({/boma x})$.

In practice, the structural function, covariance or variogram, is only used for limited distances $/vert{/boma h}/vert / < b$. The limit $b$ represents, for example, the diameter of the neighborhood of estimation (i.e., the zone which contains the information to be used). In such a case we speak of quasi-stationarity.



Rudolf Dutter 2003-03-13