The feature transformation is the same as (5.1) in Section 5.1:
|
(7.1) |
but the similarity ends here. Because is limited to a compact set,
the unit hypercube
, an energy statistic is
theoretically not needed for MEPP (See Theorem 2 in Section 3.2.1)
7.1Whereas the exponential distribution played a central
role in the analysis of the singly-bounded case,
the mutivariate truncated exponential distributon (TED) plays an
important role in the unit hypercube
(all elements of in ).
The uniform reference distribution,
results in maximum entropy
(See Theorem 2 in Section 3.2.1), and is a special case
of the mutivariate TED. When
, the J-function
simplifies to
So, computing the J-function is primarily a task of computing
,
which can be done using two methods as described in Section 17.7. The J-function is implemented by software/module_A_dbound.m, and
tested by software/module_A_dbound_test.m.
The re-synthesis approach using UMS is described in Secion 7.2
and implemented by software/module_A_dbound_synth.m.