TABLE OF CONTENTS This is the first in a series of chapters that demonstrate application of RMM to modeling systematic variation (Chapters 14 18). In this chapter we address data-sets that have appeared in the literature and analyzed by some existing methodology for empirical relational modeling. RMM solutions derived for a sample of problems are compared to solutions obtained by current approaches.
Since detailed RMM solutions, with respective major summary results extracted from the computer's runs, have already been demonstrated in Chapter 8, the results displayed here are confined to main results and goodness-of-fit statistics. Showing only essential statistics relevant to assess the quality of the derived solutions, one may find it easier to compare the effectiveness of the solutions attained by the different methodologies.
A sample of four problems is analyzed. The first two problems, related to experimental data and to field data, are entirely new to the book. RMM solutions are developed and compared to solutions derived by other methodologies. This is the subject of Section 14.2. In the next Section 14.3 we relate to two problems, already solved by RMM in Chapter 8. The two solutions are compared to those obtained by other current methodologies.
For all four problems, we compare the linear predictor (LP), obtained by RMM, to LPs obtained by linear regression (applied to the original...
