Documentos de Académico
Documentos de Profesional
Documentos de Cultura
1. Probability
2. Value of Information
3. Sensitivity Analysis
4. Simulation
5. Decision Tree
6. Standard Deviation & Coefficient of
Variation
7. Project Beta
PROBABILITY
PROBABILITY
Decision Making under
Certainty
A probability distribution describes
the chance or likelihood of each of
the collectively exhaustive and
mutually exclusive set of events.
1. Mutually Exclusive- if two events cannot
occur simultaneously.
2. Joint Probability- that two events will
both occur.
3. Conditional Probability of two events- is
the probability that one will occur given that
the other has already occurred.
4. Independent- if the occurrence of one has
no effect on the probability of the other.
Illustrative Case 21-1.Decision
Making under Uncertainty
M & O Corporation is considering two new
designs for their kitchen utensil products – Product
A and Product B. Either can be produced using the
present facilities. Each product requires an
increase in annual fixed costs of P4, 000,000.00.
The products have the same selling price of P1,
000 and the same variable cost per unit P800.
After studying past experience with similar products,
management has prepared the following probability
distribution:
Event Probability for
(Units Demanded) Product A
Product B
5,000 0.0 0.1
10,000 0.1 0.1
20,000 0.2 0.1
30,000 0.4 0.2
40,000 0.2 0.4
50,000 0.1 0.1
1.0 1.0
Management would like to know:
a) The break-even point for each product.
State of nature =
Season’s Actual Decision = Decision =
Decision =
Demand Order 0 Order 1 Order
2
0 yacht 0 P (50,000) P (100,000)
1 yacht 0 200,000 150,000
2 yachts 0 200,000 400,000
The probabilities of the season’s demand are
Pr Demand
0.10 0
0.50
0.40 1
2
The dealer may calculate the expected value of each decision as follows:
**The decision with the greatest expected value is to order two yachts, so, in
the absence of additional information, the dealer should order two.
Perfect Information- is the knowledge that a future state
of nature will occur with certainty,ie, being sure of what
will occur in the future
Expected value of perfect information- is the difference
between the expected value without perfect information
and the return if the best action is taken given perfect
information.
Market
State
Choose Profit in Probabil EV in
(P000’s) ity (P000’s)
I A 75 0.5 37.5
II B 80 0.2 16.0
III C 90 0.3 27.0
EV of Profits, with Perfect Information
80.5
Since the EV of profits without
information is P56,500 (choosing
project C), the value of perfect
information to the company is
(P80,500 - P56,500=P24,000)and the
cost of information is P15,000, it
would be worthwhile to obtain it.
Sensitivity analysis describes
how sensitive the linear
solution is
programming optimal
to a change in any one number.
Sensitivity analysis answers
what-if questions about the effect
of change in prices or variable
cost; changes in value; addition
or deletion of constraints; and
changes in industrial coefficients.
A trial-and-error method may be adopted in which the
sensitivity of the solution to changes in any given
variable, parameter, or other assumption is
calculated.
The risk of the project being simulated may also be
estimated.
The best project may be one that is least sensitive to
changes in probalistic inputs.
A sensitivity analysis may indicate whether
expending additional resources to obtain better
forecasts of future conditions is cost justified.
In linear programming problems, sensitivity
is the range within which a constraint value,
such as a cost efficient or any other variable,
may be changed without changing the optimal
solution. Shadow price is the synonym for
sensitivity in that context.
Financial planning models, including those
for cash flows and capital budgeting are
other significant applications of sensitivity
analysis.
Illustrative Case 21-3.
Application of Sensitivity Analysis
Mirmo Company has prepared the following budgeted
profitability statement for the current year operations:
Sales (2,500 units x P40) P100,000
Variable cost:
Materials P40,000
Labor 30,000 70,000
Contribution margin 30,000
Less: Fixed cost 20,000
Profit P10,000
Required:
Make sensitivity analysis based on the above data.
Solution:
1. If selling price is reduced by more than 10% budgeted,
the company would incur loss.
2. If the sales are reduced by more than 10% of the
budgeted sales of 2,500 units, the company would incur
loss.
3. If labor cost increase by more than 33.33% above the
budgeted, the company would make a loss.
4. If material cost increases by 25% or more of the
budgeted cost, the company would make a loss.
5. If the fixed costs increase by more than 50% of the
budgeted fixed cost, the company would incur loss.
Assurance requires validation of the
model-often using historical data
4. Design the experiment
Experimentation is sampling the
operation of a system
5. Conduct the simulation- evaluation results
The simulation should be conducted
with care
The results are analyzed using
appropriate statistical methods
ADVANTAGES AND
LIMITATIONS OF SIMULATION
Advantages
Time can be compressed
Alternative policies can be explored
Complex systems can be analyzed
Limitations
Cost
Risk of error
Illustrative Case 21-4. Simulation
Technique
The financial controller of Minitoons, lnc. has
drawn the following projections with
probability distributions:
Wages and Probability Raw Probability Sales Probability
Salaries Material Revenue
(P000's) (P000's) (P000's)
12-14 .2 42-46 .6
Requirement:
Wages 2 7
9 2 9 8
and
salaries
(P000's)
Raw 4 4 1 0 3 4
materials
(000's)
Sales 0 6 6 7 0 2
revenue
(P000's)
Fixed cost=14,000/month
Solution:
a. Simulation of Cash Flow Projection
Random Number
Allocation
Wages and Salaries Raw Materials Values Revenue
Mid- Cumula Rando Mid- Cumul Rando Mid- Cumul Rando
point tive m point ative m point ative m
(P000's) probabi numbe (P000's probab numbe (P000's probab numbe
lty rs ) ilty rs ) ilty rs
11 .3 0-2 7 .2 0-1 32 .1 0
Decision 2
Success Competitors come in
Postpone
Launch
Continu Competitors stay out
e Project
Immediately Launch
Abort
Project Competitors come in
Competitors stay
out
Illustrative Case 21-5. Decision
Tree
with Qualified Outcomes.
Immediately
Launch
Competitors come in
Competitors come in
Abort
Project
Competitors come out
1. It will cost an estimated P50,000 to continue the
project which itself probabilistic.