Consider an arbitrary distribution . Our approach consists of two steps, the first of which involves constructing a baseline EC distribution, and the second of which involves reducing the number of phases in this baseline solution. If , then the baseline solution used is simply given by the Simple solution. Also, if but , then it turns out that the Simple solution could be defined for this , and this gives the baseline solution. If but , then to obtain the baseline EC distribution we first find a distribution such that and and then set such that is well-represented by distribution , where (see Figure 2.14(a)). The parameters of the EC distribution that well-represents are then obtained by the Simple solution.
|
To reduce the number of phases used in the baseline EC distribution, we exploit the subset of two-phase Coxian PH distributions that are not used in the Simple solution. The Simple solution is based on the fact that a distribution is well-represented by a two-phase Coxian PH distribution when . In fact, a wider range of distributions are well-represented by the set of two-phase Coxian PH distributions. In particular, if is in set , then is well-represented by a two-phase Coxian PH distribution (see Figure 2.14(a)). By exploiting two-phase Coxian PH distributions in , the Complete solution reduces the number of phases used. The above ideas lead to the following solution:
(i) If , then the Simple solution provides the parameters (, , , , ).
(ii) If
(see Figure 2.14(a)),
where
denotes the complement of ,
then let
(iii) If
,
then the Simple solution provides the
parameters (, , , , ), except that
(2.6) is replaced by
Proof:By Theorem 3,
is a continuous and monotonically increasing function of ,
Thus,