Model choice and adjustment techniques in the presence of prior information
Loading...
Date
1983-09
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Ohio State University. Division of Geodetic Science
Abstract
The frequent problem of estimating nonidentifiable parameters within a singular Gauss-Markov model may be overcome by introducing some independent prior information, in addition. In this case, we are able to improve either the model or the estimator (or both simultaneously), which leads us to modified estimation and/or prediction formulas. These improved adjustment techniques are investigated in more detail with respect to (weak or uniform) unbiasedness, minimum variance, minimum mean square error, etc. The determination of the earth's crustal deformations from repeated baseline measurements using geophysical prior information may well serve as an example for applications.
Description
Prepared for National Aeronautics and Space Administration, Goddard Space Flight Center, Greenbelt, Maryland: Grant No. NSG 5265, OSURF Project 711055