What do you think?
Rate this book
Decision Analysis: Introductory Lectures on Choices Under Uncertainty
The approach taken in this book prescribes how an individual who is faced with a problem of choice under uncertainty should go about choosing a course of action that is consistant with his personal basic judgments and preferences. The author has minimized the mathematical demands upon the reader and most of the book is free from algebraic and symbolic manipulations. Sophistication in logical analysis is required throughout.
- GenresNonfictionMathematics
309 pages, Paperback
First published June 1, 1968
3 people are currently reading
66 people want to read
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 3 of 3 reviews
May 20, 2023
Several years ago I took a class with Richard Zekchauser that, in short, helped reformulate a lot of the way I think about uncertainty in the world. For example, imagine trying to find the answer to a question "how many homeless children live in Katmandu," or, imagine you were grading a 50 question multiple choice test for 20 students, but didn't have the grading rubric? How would you most accurately grade the test? Everything is about uncertainty and cleverness, and this book was written by Richard's mentor Howard Raiffa - a/the father of decision analysis and an early (aggressive) proponent of Bayesian perspectives in statistics. This book was SUPERB if and only if you want an intense mathematical dive into decision analysis (you probably don), but if you do READ!!!
A few things I learned.
1. The practice of outlining subjective probability assessments of a given scenario is both useful, and a skill, and people are SHOCKINGLY bad at it. I've been trying to make 90% confidence around things recently and I miss SO much (wayyy more than 10% of the time)
2. Contemplating if more information actually changes a decision is a difficult, yet highly useful practice. No need to keep searching if info changes nothing!
3. One can ALWAYS randomize across a set of decisions as a convext combination of probabilities to get a desired expectation that is along a similar utility line. This usually doesn't change a given payoff for a choice in expectation for a rational agent, but if someone has different risk tolerances, it can allow for more optionality in a given situation.
A few things I learned.
1. The practice of outlining subjective probability assessments of a given scenario is both useful, and a skill, and people are SHOCKINGLY bad at it. I've been trying to make 90% confidence around things recently and I miss SO much (wayyy more than 10% of the time)
2. Contemplating if more information actually changes a decision is a difficult, yet highly useful practice. No need to keep searching if info changes nothing!
3. One can ALWAYS randomize across a set of decisions as a convext combination of probabilities to get a desired expectation that is along a similar utility line. This usually doesn't change a given payoff for a choice in expectation for a rational agent, but if someone has different risk tolerances, it can allow for more optionality in a given situation.
October 19, 2009
It's probability, without all the stupid notation! Apparently Raiffa was quite controversial at the time...on the other hand, he's the dude who got his PhD as soon as he turned in his orals presentation (because it was the level of a dissertation already).
April 26, 2016
A good book for the basics, I think/hope. Slightly more mathematically demanding than the book by Lindley. As the non-expert, it is a bit hard to ssy what is still relevant, but at least the
Displaying 1 - 3 of 3 reviews