Decision Analysis: Decision Trees, Expected Value & Monte Carlo
Payoff tables and decision criteria, decision-tree roll-back, the value of perfect and sample information with Bayes, utility and certainty equivalents, Monte Carlo NPV, tornado sensitivity and real options: 8 graded labs on one robotics company's decisions.
The first lesson is free — no account needed.
About this course
A decision analysis course that works through one robotics company's product decisions in a spreadsheet, for managers, analysts and MBA students. You compare three ways to build a robot in a payoff table with maximax, maximin, minimax regret and expected value, roll back a two-stage decision tree, price a perfect demand forecast with EVPI and check it against the expected opportunity loss, update probabilities with Bayes' rule to value a market survey, find certainty equivalents with an exponential utility function, turn given random numbers into 30 futures of a project NPV to read the expected value and the chance of a loss, rank the drivers in a tornado chart from P10 to P90 and value the options to wait and to expand. The first lab is free.
Lessons
- Payoff Tables & Decision Criteria: Maximax, Maximin, Regret and EVFree
- Decision Trees: Rolling Back Chance and Decision Nodes Pro
- Value of Perfect Information (EVPI) Pro
- Value of Sample Information: Bayes Update of a Market Survey Pro
- Risk Attitude & Utility: Certainty Equivalents with Exponential Utility Pro
- Monte Carlo Simulation of a Project NPV Pro
- Sensitivity Tornado: Which Uncertainty Drives the NPV? Pro
- Real Options: The Value of Waiting and of Expanding Pro