Aa. Grimstad et al., Assessing the validity of a linearized accuracy measure for a nonlinear parameter estimation problem, INVERSE PR, 17(5), 2001, pp. 1373-1390
We consider accuracy assessment for the inverse problem of recovery of unkn
own coefficient functions in differential equations from data containing ra
ndom errors. The set of PDEs constituting the current forward model describ
es a special case of two-phase porous-media flow. We are concerned mainly w
ith two issues. (1) When is it valid to calculate parameter accuracies for
the current nonlinear estimation problem by a linearized method, linearized
covariance analysis (LCA)? (2) Can the validity of LCA be assessed without
performing an accurate, but computationally very expensive, Monte Carlo an
alysis (MCA)? For both issues, special emphasis is put on parameter subsets
for which LCA predicts high accuracy.
The curvature measures of nonlinearity (CMNs) are a potential alternative t
o MCA. CMNs are approximate, but considerably less expensive to compute. In
this paper, we apply LCA, CMNs and MCA to several instances of the Current
model. We address issue I by comparing LCA and MCA results, and issue 2 by
including also CMN results in the analysis. It is found that CMN and MCA r
esults lead to identical and negative conclusions concerning the validity o
f LCA.
However, if the real concern is parameter subsets where LCA predicts high a
ccuracy, these conclusions, based on calculations involving all of the para
meters, were often misleading. Use of specially designed subset CMNs is ess
ential to avoid this. A potential explanation, which may have implications
also for other parameter estimation problems, is presented.