A Treatise On Probability (Paperback)

by John Maynard Keynes (Author)
Key Phrases: inductive correlation, urn containing black, statistical induction, Principle of Indifference, Von Kries, Laws of Thought (more...)

Editorial Reviews

Product Description
An Unabridged, Digitally Enlarged Printing: The Meaning Of Probability - Probability In Relation To The Theory Of Knowledge - The Measurement Of Probabilities - The Principle Of Indifference - Other Methods Of Determining Probabilities - The Weight Of Arguments - Historical Retrospect - The Frequency Theory Of Probability - The Theory Of Groups, With Special Reference To Logical Consistence, Inference, And Logical Priority - The Definitions And Axioms Of Inference And Probability - The Fundamental Theorems Of Necessary Inference - The Fundamental Theorems Of Probable Inference - Numerical Measurement And Approximation Of Probabilities - Some Problems In Inverse Probability, Including Averages - The Nature Of Argument By Analogy - The Value Of Multiplication Of Instances, Or Pure Induction - Some Historical Notes On Induction - The Meanings Of Objective Chance, And Of Randomness - Some Problems Arising Out Of The Discussion Of Chance - The Application Of Probability To Conduct - The Nature Of Statistical Inference - The Law Of Great Numbers - The Theorems Of Bernoulli, Poisson, And Tchebycheff , etc., etc. - Bibliography And Comprehensive Index

From the Publisher
--This text refers to the Paperback edition.

Product Details

• Paperback: 480 pages
• Publisher: Rough Draft Printing (June 19, 2008)
• Language: English
• ISBN-10: 1603861181
• ISBN-13: 978-1603861182
• Product Dimensions: 9.3 x 7.5 x 1.1 inches
• Shipping Weight: 1.8 pounds (View shipping rates and policies)
• Average Customer Review: 3.7 out of 5 stars See all reviews (7 customer reviews)
• Amazon.com Sales Rank: #56,065 in Books (See Bestsellers in Books)

  Popular in these categories: (What's this?)

  #2 in	Books > Professional & Technical > Professional Science > Mathematics > Pure Mathematics > Algebra > Abstract
  #2 in	Books > Science > Mathematics > Pure Mathematics > Algebra > Abstract
  #22 in	Books > Science > Experiments, Instruments & Measurement > Methodology & Statistics

Customer Reviews

Most Helpful Customer Reviews

35 of 43 people found the following review helpful:

5.0 out of 5 stars The best book published on the foundations of probability, May 16, 2005

By	Michael Emmett Brady "mandmbrady" (Bellflower, California ,United States) - See all my reviews (TOP 1000 REVIEWER)    (REAL NAME)

This review is from: A Treatise on Probability (Phoenix Edition) (Hardcover)

In this path breaking contribution to the logic of probability,Keynes showed how to adapt the work of George Boole for the purpose of estimating probabilities.Keynes is the first scholar in history to explicitly emphasize the importance of interval estimates in decision making.For Keynes there are only two types of probability estimates,point estimates and interval estimates.Unfortunately,Keynes decided to call interval estimates "non-numerical"probabilities.His reasoning is really quite obvious.A precise estimate of probability used a single numeral for the point estimate.Therefore,an imprecise estimate of probability used two numerals to denote an interval(set).Thus, an interval estimate is not based on a single numeral but two. These types of probabilities are thus "non-numerical"because you are not using a single numeral.In 1922 and 1926,Frank Ramsey reviewed Keynes's book based on his reading of chapters 1-4 plus 3 pages from Part two and 4 pages from Part five.Keynes's discussion of non-numerical probabilities takes place in chapters 5,10,15 and 17.Keynes then applies his new approach to induction and analogy in chapters 20 and 22,using his concept of "finite probability"which applies to both precise numerical probabilities and imprecise non-numerical probabilities.All of Keynes's discoveries ,however,were ignored by the ignorant Ramsey.(To this day(2005)one can regularly read about Keynes's "strange,mysterious,unfathomable,undefined"non-numerical probabilities in literally hundreds of economics and philosophy journal articlesthat are based primarily on Ramsey's reviews.These reviews are still cited as "overwhelming" evidence that Keynes agreed that Ramsey's critique had demolished the entire structure of his logical approach to probability.Nothing could be further from the truth.Ramsey's reviews were so poor that Keynes and Bertrand Russell attempted to downplay their relevance so as to save Ramsey from being embarrassed in the academic community.)Keynes then showed that interval estimates,because they overlap,would very likely also,in many cases,be noncomparable and/or nonrankable if a decision maker used such order preserving operators like"greater than or equal to"or "less than or equal to".While this is quite obvious,it went completely over Ramsey's head. Keynes's second major advance was to create his "conventional coefficient of weight and risk", c=p/(1+q)[2w/(1+w)]. The goal of the decision maker is to Maximize cA,where A is some outcome.This decision rule solves all of the paradoxes and anomalies that plague subjective expected utility theory.A major accomplishment made by Keynes in chapter 26 of the TP was to specify that the weight of the evidence variable,w,was defined on the unit interval [0,1].It would be forty years before Daniel Ellsberg would define his practically identical variable,rho,on the unit interval between 0 and 1 also,where rho measured the degree of confidence in the decision maker's information base.Since these two measures are one to one onto and isomorphic,Keynesian weight(uncertainty in the General Theory) and Ellsbergian ambiguity measure the same thing and are interchangeable.This means that Ellsberg's analysis can be applied when studying the GT and used to buttress Keynes's theory of liquidity preference in the GT.In Part 5 of this book ,Keynes showed how one could use Chebyshev's Inequality as a lower bound to the normal probability distributions overly precise point estimate . Part 5 of the Treatise also includes Keynes's advocacy of the Lexis Q test for stability of a statistical frequency[law of large numbers].It is this part of the TP that forms the basis,along with chapter 17,of Keynes's exchange with Tinbergen over the logical foundations of econometrics in the Economic Journal in 1939-1940.Keynes pointed out that ,in order to justify his assumption of normality,Tinbergen needed to apply the Lexis Q test.Tinbergen never applied either that test or the Chi- Square test for goodness of fit.This will then bring the reader back to Keynes's chapter 8 of the Treatise ,where he presents his own logical frequency interpretation of probability as a special case of his general logical approach to probability.

6 of 9 people found the following review helpful:

3.0 out of 5 stars Historically important ... but., November 2, 2008

By	J. Hall (El cajon, ca United States) - See all my reviews (REAL NAME)

The author introduces key concepts for the first time. He also introduces some ideas that I had not seen before. The book is historically important. However, the author's writing style is very much a product of his time - obviously brevity was not considered a virtue. The book is really, painfully long. If you are a historian, great. If you are looking for key concepts in statistical theory, there are other books that will get to the point much quicker.

11 of 17 people found the following review helpful:

5.0 out of 5 stars Treats mathematical probability as a limiting case of logical probability, September 18, 2006

By	Michael Emmett Brady "mandmbrady" (Bellflower, California ,United States) - See all my reviews (TOP 1000 REVIEWER)    (REAL NAME)

This review is from: A Treatise on Probability (Paperback)

In this path breaking contribution to the logic of probability,Keynes showed how to adapt the work of George Boole for the purpose of estimating probabilities.Keynes is the first scholar in history to explicitly emphasize the importance of interval estimates in decision making.For Keynes there are only two types of probability estimates,point estimates and interval estimates.Unfortunately,Keynes decided to call interval estimates " non-numerical "probabilities.His reasoning is really quite obvious.A precise estimate of probability used a single numeral for the point estimate.Therefore,an imprecise estimate of probability used two numerals to denote an interval(set).Thus, an interval estimate is not based on a single numeral but two. These types of probabilities are thus " non-numerical "because you are not using a single numeral.In 1922 and 1926,Frank Ramsey reviewed Keynes's book based on his reading of chapters 1-4 plus 3 pages from Part II and 4 pages from Part V.Keynes's discussion of non-numerical probabilities takes place in chapters 5,10,15 and 17.Keynes then applies his new approach to induction and analogy in chapters 20 and 22,using his concept of " finite probability " ,which applies to both precise ,numerical probabilities and imprecise, non-numerical probabilities.All of Keynes's discoveries ,however,were ignored by the ignorant Ramsey.It is unfortunate that the editorial foreword to the 1973 Collected Writings of JMK edition of the TP, written by Richard Braithwaite ,simply repeats all of the errors made by Ramsey in his reviews.Consider Braithwaite's paraphrase of Ramsey's argument that " On Keynes's theory it is something of a mystery why the probability relations should be governed by the probability calculus."(p.xx,1973).The answer is quite simple. First,the " non numerical " interval estimates will not be governed by the probability calculus.Second,numerical probability calculations,such as the blue-green taxi cab problem of Tversky and Kahneman,will only satisfy the probability calculus if the weight of the evidence,w,is equal to 1,where w is defined as an element on the unit interval between 0 and 1 and measures the relative completeness of the available evidence upon which the probability estimates are to be calculated.(To this day(2006)one can regularly read about Keynes's " strange,mysterious,unfathomable,undefined " non-numerical probabilities in literally hundreds of economics and philosophy journal articles and books that have been written about Keynes's approach to probability since the Ramsey reviews were first published 80 years ago.These reviews are still cited as " overwhelming " evidence that Keynes agreed that Ramsey's critique had completely demolished the entire structure of his logical approach to probability.Nothing could be further from the truth.Ramsey's reviews were so poor that Keynes and Bertrand Russell attempted to downplay their relevance so as to save Ramsey from being embarrassed in the academic community.)Keynes then showed that interval estimates,because they frequently overlap,would very likely also,in many cases,be nonmeasurable,noncomparable and/or nonrankable if a decision maker used such order preserving operators like " greater than or equal to " or "less than or equal to ".While this is quite obvious to any reader of Part II of the TP,it went completely over Ramsey's head. Keynes's second major advance was to create his "conventional coefficient of weight and risk", c=p/(1+q)[2w/(1+w)] in sections 7 and 8 of chapter 26 . The goal of the decision maker is to Maximize cA,where A is some outcome.This decision rule solves most of the paradoxes and anomalies that plague subjective expected utility theory.A major accomplishment made by Keynes in chapter 26 of the TP was to specify that the weight of the evidence variable,w,was defined on the unit interval [0,1].It would be forty years before Daniel Ellsberg would define his practically identical variable,rho,on the unit interval between 0 and 1 also,where rho measured the degree of confidence in the decision maker's information base.Since these two measures are one to one onto and isomorphic,Keynesian weight(uncertainty in the General Theory) and Ellsbergian ambiguity measure the same thing and are interchangeable.This means that Ellsberg's analysis can be applied when studying the GT and used to buttress Keynes's theory of liquidity preference in the GT.In Part 5 of this book ,Keynes showed how one could use Chebyshev's Inequality as a lower bound to the normal probability distributions overly precise and inaccurate point estimate . Part 5 of the Treatise also includes Keynes's advocacy of the Lexis Q test for stability of a statistical frequency[law of large numbers].It is this part of the TP that forms the basis,along with chapter 17,of Keynes's exchange with Tinbergen over the logical foundations of econometrics in the Economic Journal in 1939-1940.Keynes pointed out that ,in order to justify his assumption of normality,Tinbergen needed to apply the Lexis Q test.Tinbergen never applied either that test or the Chi- Square test for goodness of fit.This will then bring the reader back to Keynes's chapter 8 of the Treatise ,where he presents his own logical frequency interpretation of probability as a special case of his general logical approach to probability after criticizing Venn's particular version of a frequency approach.