Next: Matrix Form
Up: The Kriging Estimator
Previous: Non-bias Condition
  Contents
Minimum Estimation Variance
The estimation variance
can be expanded as follows:
with
The terms
are eliminated and, thus, we have
The estimation variance can, thus, be expressed as a quadratic form in
, and is to be minimized subject to the non-bias condition
.
The optimal weights are obtained from standard Lagrangian techniques by
setting to zero each of the
partial derivatives (in respect to
and
to the Lagrange parameter
) of
This procedure provides a system of
linear equations in
unknowns (the
weights
, and the Lagrange parameter
) which is called the ``kriging system'':
The minimum estimation variance, or ``kriging variance'', can then be
written as
This kriging system can also be expressed in terms of the semi-variogram
function
, particularly when the random function
is intrinsic only and the covariance function
is not defined:
Next: Matrix Form
Up: The Kriging Estimator
Previous: Non-bias Condition
  Contents
Rudolf Dutter
2003-03-13