Comparing uncertain strategies

Compare uncertain strategies using probabilities payoffs and model assumptions

Problem types

Using expected value to compare choices
Finding a conditional probability
The multiplication rule for probability · Useful idea
Along a decision tree, probabilities multiply along each path to find the chance of each payoff.
A distribution from a sample space · Useful idea
Expected values computed from a theoretical distribution are the averages used to compare uncertain choices.
Quadratic, rational, and exponential constraints · Useful idea
Finding the best choice often means finding the best value of a model, such as the maximum of a quadratic.
Reading a statistical report critically · Useful idea
A decision rests on the evidence reported, so a report is critiqued for flaws before it is used.
Designs that meet geometric constraints · Useful idea
Choosing the best design among options compares their values, as a decision strategy compares expected values.
Income tax, budgets, and net worth · Useful idea
Choosing between incentives or plans compares expected values, which is a decision strategy.

A distribution built from frequencies · See it applied
When expected value is hard to compute, a decision strategy is evaluated by simulating it and averaging the simulated results.

A procedure that gives each outcome a fair chance · Related idea
A decision that picks among options by a random device needs the device to be fair.
Checking a model against evidence · Related idea
A decision made by expected value is only as good as the probabilities in the model, which is why a model is checked against data.
Independence and mutual exclusion · Shared foundations

Open the interactive map to explore lessons and skill connections.