ADM 2302 Lecture Notes - Lecture 18: Emv, Backward Induction, Statistical Hypothesis Testing
6 Jan 2018
School
Department
Course
Professor
Document Summary
> replacing these nodes, we work backwards from the end of the tree, computing expected utilities and decision values as we go, until we replace the entire tree with a single leaf. > in replacing the decision nodes, we need to record which decisions achieve the maximum these are the decisions we want to take if we ever reach this node. = ,000 x 0. 30 + ,000 x 0. 50 + ( ,000) x 0. 20 = ,000. Emv (node 3) = emv (no plant) = sh. Emv (node 4) = emv (large plant | positive result) = ,000 x 0. 509 + ,000 x 0. 468 + ( ,000) x 0. 023 = ,840. Emv (node 5) = emv (small plant | positive result) = ,000 x 0. 509 + ,000 x 0. 468 + ( ,000) x 0. 023. Emv (node 6) = emv (no plant | positive result) Emv (node 7) = emv (large plant | negative result)
