Thinking and Deciding

Rating:
Summary: The logical theory of probability is not the classical th.
Review: This book is badly marred by the author's failure to carefully differentiate between the classical theory of probability,which was founded in the 17th century,and the logical theory of probability,which J M Keynes started early in the 20th century.Rudolph Carnap made additional contributions in the mid-1950's to the logical theory of probability. In this reviewer's opinion,pp.93-103 contain so many errors of both ommission and commission that a novice reader will come away from reading this book with serious gaps in his understanding of what probability means and how it is applied in the real world.The fact that Baron is a strong supporter of the Ramsey-de Finetti subjectivist approach does not justify the very misleading appraisels of the relative frequency and logical theories of probability in his book. Contrary to Baron,Carnap's approach makes use of all the available evidence,including statistical data and relative frequencies.Carnap's approach also makes use of the betting quotients technique used by subjectivists.It simply is not true that"Likewise,the logical view is ordinarily useless as a justification for making probability judgments,except in textbook(or casino)cases of"fair coins"....".The second major draw back of this book is the author's strong belief that all probabilities are precise,exact,definite point estimates.Baron apparently believes that there is always sufficient information to represent the probability by a unique single number . (see Baron,pp.94-96). Baron appears to be completely ignorant of J M Keynes's interval valued approach or of the work of Walley ,Levi,Kyburg and The Society for Interval Valued Probability Theory and Application(SIPTA) in reviving Keynes's long dormant approach.

Rating:
Summary: Decisions, decisions, decisions
Review: This is a _very_ impressive book. I can't imagine a more thorough overview of the study of human judgment, decisionmaking, and rationality.

It's hard to find a good single volume in this field. For the psychology of judgment and decisionmaking, there's Scott Plous's excellent book of that title. For the heuristics on which we seem to rely and the biases they seem to generate, there's the modern classic _Judgment Under Uncertainty_ by (the late) Amos Tversky and (recent Nobel prize winner) Daniel Kahneman. And there are other books devoted to special topics and subtopics.

But so far as I know, this is the only broad, general introduction to the entire field of thinking, how we do it, how we probably ought to do it, and the sorts of things we do it about.

The book is divided into three broad sections. "Thinking In General" covers just that: what thinking is, and the nature of rationality and logic. "Probability and Belief" introduces not only probability theory (including, importantly, Bayes's Theorem) but the various theories about what probability _is_, as well as hypothesis testing, correlation, and "actively open-minded thinking." The largest section, "Decisions and Plans," is eleven chapters long and covers everything else: uncertainty, utility, decision analysis, theories of morality fairness, and justice, and risk.

As you might imagine, Baron doesn't leave too many stones unturned. Heck, his bibliography alone is thirty-odd pages long.

The book is eminently readable. The third edition is aimed mainly at graduate students but I gather Baron still uses it in his undergraduate courses. If you're a lay reader interested in this field, you'll be able to read it on your own with no difficulty.

Rating:
Summary: A great introduction to Decision Theory
Review: This is the first book I recommend to anyone interested in learning how to think and decide better. It is both an exploration of the usual way we think (and the mistakes we make) as well as a contruction of a toolkit for better thinking. It's well written and is at the level of an advanced undergraduate liberal arts course. Not too technical and mathematics oriented, but covers the concepts with rigor in an intuitive way. A good preparation for more technically oriented investigations of the subject.