Complexity, Risk, and Financial Markets 1st Edition

very useful for non mathematicians: the author succeeds in passing intuition of the lot of abstract concepts defining chaos theory and complexity theory

Peters does an excellent job in clearly differenciating between the concepts of complexity, ignorance,risk,uncertainty,vagueness,and ambiguity.He shows how each concept has an important separate ,yet interrelated ,clearcut role to play in financial decision making.Unfortunately,Peters overlooks the fact that John Maynard Keynes had already provided an operational approach for such a decision making approach in his A Treatise on Probability(TP)in 1921.The role of conflicting and ambiguous evidence in decision making in general was discussed by Keynes in his rain-barometer-dark clouds example in chapter 3 of the TP.Keynes's point in discussing this particular problem was to show that it would be difficult to measure and quantify this particular dimension of a decision problem.The implicit danger is that no account will be taken of it if a strictly quantitative approach to decision making is taken.The importance of vagueness was discussed by Keynes in chapter 1,p.5,chapter 2,p.17 and in chapter 22,p.259.Complexity is discussed within the context of Keynes's discussion of induction and analogy.The ability to reason inductively starts to break down if the number of important independent factors(Keynes's principle of limited(finite)independent variety)grows too large and/or the interactive effects start to exhibit nonlinear relationships instead of proportionate,linear relationships.The clearcut differences between ignorance,uncertainty and risk in general decision making were discussed in chapters 6(the introduction)and chapter 26(the conclusion).Keynes defined an index to measure what he called the weight of the evidence,w, upon which probability estimates would be based.w is a measure of the completeness and the reliability of the relevant,potential,available evidence and information that a decision maker can obtain( or be aware of its existence).w is defined on the unit interval [0,1],where 0<=w<=1.Keynes's discussion of the Ellsberg Paradox takes place in chapter 6 of the TP forty years before Ellsberg published his 1961 paper.Keynes's w is practically the same as Ellsberg's rho measure of ambiguity.In the General Theory,Keynes defines uncertainty to be an inverse function of the weight of the evidence,where u(uncertainty)is a function of w.Thus,u=f(w).Ignorance(w=0),uncertainty(0<w<1)and risk(w=1)are all defined by Keynes relative to his w index.Risk can then be either linear(proportionate) or nonlinear(nonproportional).Keynes wraps all of this up into his conventional coefficient of weight and risk,c. c is a decision rule that incorporates nonlinearities in decision making.The goal of the decision maker is to maximize cA,where A is some outcome and c equals (p/1+q)(2w/1+w).This rule generalizes the expected value rule and the expected utility rule.Keynes doesn't make the common error of conflating risk with diminishing marginal utility.Finally,Keynes is the founder of the interval estimate approach to probability.Each probability has an upper and a lower bound .Keynesian expected values are thus also intervals.Keynes's development of an interval estimate approach to the estimation of probabilities in chapters 15 and 17 of the TP HAS BEEN OVERLOOKED DUE TO THE INFLUENCE OF TWO ERROR FILLED BOOK REVIEWS WRITTEN BY FRANK RAMSEY IN 1922 AND 1926,respectively.Ramsey completrly misinterpreted Keynes's definition of the words"nonnumerical" and"nonmeasurable".Keynes uses them to mean that two numerals ,not one, are needed to specify probabilities in general.Ramsey mistakenly assumed that Keynes was arguing that numbers could not be used in general to quantify the probability relation.This reviewer concludes that Peters should have based his overall approach upon Keynes's work and not upon an Austrian approach to economic theory which is impossible to operationalize due to their bias against any kind of quantification.Ramsey's criticisms are correct if applied to the Austrian position on the use of quantitative methods in decision theory.They are off target when applied to Keynes.A revised version of Peter's book would be even more convincing if he incorporated a chapter on Keynes's nearlty 80 year old approach.

I'm not a math gueek, evolutionist, or statistician. What I wanted was someone to expalain, in as simple terms as possible, what "complexity" is in terms of systems analysis - be that social, economic or otherwise. Kudos to Peters for delivering just that. I read the negative reviews written before my own, considered them, and then I bought the book anyway.Frankly, I'm glad Peters "dumbed down" this topic for me. That's what I wanted. Not knowing much about the subject, I was still able to sense where some errors of omission were being made. And there are a few. I wish he would have finished his thought on the Monte Hall problem, for example. Don't buy this book if you are thinking it's full of exciting mathematical equations or advanced theory. It isn't. It's just an introduction to complex/chaotic systems written for the average Joe/Jane.For all the flap over "missing Keynes' contribution" or "misrepresenting Darwin," he may very well do that. I did not buy this book expecting Peters to be a Keynes scholar or a Darwinian evolutionist. If he has read 0 Keynes and 0 Darwin, I think we can still consider him an expert on complex and chaotic systems. Certainly, Keynes and Darwin were not. I expected that Peters might know a little bit about chaos theory and complexity as it relates to the realm of economics. And that he does. So, again, kudos to Mr. Peters for dumbing down this complex topic (pun intended) enough to create a starting point for me.