Exam 9: Decision Analysis

Figure The operations manager for a local bus company wants to decide whether he should purchase a small,medium,or large new bus for his company.He estimates that the annual profits (in $000)will vary depending upon whether passenger demand is low,moderate,or high,as follows. -If he uses Bayes' decision rule,which size bus will he decide to purchase?

What is the expected payoff from selecting script #2?

Figure The construction manager for ABC Construction must decide whether to build single family homes,apartments,or condominiums.he estimates annual profits (in $000)will vary with the population trend as follows: -If he uses Bayes' decision rule,which kind of dwellings will he decide to build?

What is the probability that script #1 will be a success,but its sequel will not?

Figure -The Bayes' decision rule strategy is:

What is the probability that the statistics book would wind up being placed with a smaller publisher?

An example of maximax decision making is a person buying lottery tickets in hopes of a very big payoff.

The maximin approach involves choosing the alternative that has the "best worst" payoff.

Determining the worst payoff for each alternative and choosing the alternative with the "best worst" is the criterion called:

A posterior probability is a revised probability of a state of nature after doing a test or survey to refine the prior probability.

Figure -The maximin strategy is:

The Sensit Spider graph is useful for examining the impact of changes in several data values at once.

Figure -The maximax strategy is:

What is the expected annual profit for the dwellings that he will decide to build using Bayes' decision rule?

The equally likely criterion assigns a probability of 0.5 to each state of nature.

Figure -The maximum likelihood strategy is:

Figure There is an option of paying $100 to have research done to better predict which state of nature will occur.When the true state of nature is S1,the research will accurately predict S1 60% of the time.When the true state of nature is S2,the research will accurately predict S2 80% of the time. -Given that the research is not done,what is the expected payoff using Bayes' decision rule?

The maximax approach is an optimistic strategy.

Prior probabilities refer to the relative likelihood of possible states of nature.

A decision tree branches out all of the possible decisions and all of the possible events.