Decision Analysis lecture 6 Flashcards
Decision Making is based on what
based hard evidence not intuition
what is Business Analytics
Applying advanced analytical methods on data to decipher future and help make better decisions in practice
what are Descriptive Analytics
Understanding past events
what are Predictive Analytics:
Predicting future events
what are Prescriptive Analytics:
Prescribing current/future decisions
these analytics are all about providing advice.
Attempt to quantify the effect of future decisions in order to advise on possible outcomes before the decisions are actually made
Prescriptive analytics allows users to do what
“prescribe” a number of different possible actions to and guide them towards a solution
what methods are used for prescriptive analytics
payoff tables/trees and decision criteria
what are the different success criterion
Maximin criterion
Maximax criterion
Maximum likelihood criterion Expected value criterion
wat is Maximin criterion:
alternative with highest guaranteed payoff
is a Maximin criterion a good way to make decisions
sometimes referred to as “the worst-case” decision making
what is Maximax criterion:
alternative with highest possible payoff (best worst case)
what is Expected value criterion:
alternative with highest expected value
this is the one you use a formula for
Expected value criterion is sometimes referred to as what
the risk neutral criterion
why Expected Value
Law of large numbers
what is the Law of large numbers
The average value of a sequence of independent and identically distributed random variables X1, X2, X3, . . . converges (“with virtual certainty”) to the expected value of the random variables
give a summary of Maximin/maximax criterion:
overly pessimistic/optimistic (focus on worst/best case) ignores all information about other cases
what is the summary of Maximum likelihood criterion:
focus on most likely case intuitively appealing but also ignores all information about other cases
what is summary of Expected value criterion:
typically considered the criterion of choice in practice
However: expected value criterion requires (rough) knowledge of all states and their occurrence probabilities!
=⇒ use sensitivity analysis to assess parameter importance!
Benefits of decision trees
Clarity of the problem:
Insight into the decision process
importance of key data
explain Clarity of the problem:
Decision trees reveal the interplay of decisions and
uncertain events over time
explain Insight into the decision process:
Decision trees show the optimal strategy, as well as what determines that strategy
explain Importance of key data:
Sensitivity analysis allows us to determine which data is most relevant for us. This often suggests where to invest more effort into data gathering.
what are some Extensions for the decision trees
Incorporation of risk attitude
Non-quantifiable consequences
explain Incorporation of risk attitude
(in particular of risk aversion) =⇒ use Utility Theory
explain Non-quantifiable consequences
(strategic benefits, “soft factors”) =⇒ use Multi-Criteria Decision Analysis
go over how to make a decision tree
do it with practice problems
