
Contents vii
5.6 The Simple Gaussian Case with a Linear Theory 91
5.7 The General Linear, Gaussian Case 92
5.8 Equivalence of the Three Viewpoints 95
5.9 The
F
Test of Error Improvement Significance 96
5.10 Derivation of the Formulas
of
Section 5.7 97
6
NONUNIQUENESS AND LOCALIZED AVERAGES
6.1 Null Vectors and Nonuniqueness
101
6.2 Null Vectors of a Simple Inverse Problem 102
6.3 Localized Averages of Model Parameters 103
6.4 Relationship to the Resolution Matrix 104
6.5 Averages versus Estimates 105
6.6 Nonunique Averaging Vectors and A Priori
Information 106
7
APPLICATIONS
OF
VECTOR SPACES
7.1
7.2
7.3
7.4
7.5
7.6
7.7
7.8
7.9
Model and Data Spaces 109
Householder Transformations
111
Transformations That Do Not Preserve Length
The Solution of the Mixed-Determined Problem I18
Singular-Value Decomposition and the Natural Generalized
Inverse 119
Derivation of the Singular-Value Decomposition
Simplifying Linear Equality and Inequality
Constraints 125
Inequality Constraints 126
Designing Householder Transformations 115
117
124
8
LINEAR INVERSE PROBLEMS AND NON-GAUSSIAN
DISTRIBUTIONS
8.1
L
,
Norms and Exponential Distributions 133
8.2
8.3 The General Linear Problem I37
8.4 Solving
L,
Norm Problems 138
8.5 TheL-Norm 141
Maximum Likelihood Estimate of the Mean of an Exponential
Distribution 135