Optimalf 173
Note that we have created our own binned distribution in creating our
scenarios here. Similarly, if we know the distributional form of the data,
we can use that and the probabilities associated, with that distribution with
this technique for finding the optimalf.Such techniques we call “paramet-
ric techniques,” as opposed to the “Empirical Techniques” described prior
to this section in the text. The Scenario Planning Approach, as described
here, where we create the data bins from empirical data, being therefore
a hybrid approach between an empirical means of determining optimalf
and a parametric one.
Scenario Spectrums
We now must become familiar with the notion of ascenario spectrum.A
scenario spectrum is a set of scenarios, aligned in succession, left to right,
from worst outcome to best, which range in probability from 0% to 100%. For
example, consider the scenario spectrum for a simple coin toss whereby we
lose on heads and win on tails, and both have a .5 probability of occurrence
(Figure 4.14).
A scenario spectrum can have more than two scenarios—you can have
as many scenarios as you like (see Figure 4.15).
This scenario spectrum corresponds to the following scenarios, taken
from the previous section pertaining to XYZ Manufacturing Corporation’s
assessment of marketing a new product in a remote little country:
Scenario Probability Result Prob×ResultWar .1 −$500,000 −$50,000
Trouble .2 −$200,000 −$40,000
Stagnation .2 $0 $0
Peace .45 $500,000 $225,000
Prosperity .05 $1,000,000 $50,000
Sum 1.00 Expectation $185,000FIGURE 4.14 Scenario spectrum for a simple coin toss in which tails wins