Skip to main content
SpringerLink
Log in

Defensive Forecasting: How to Use Similarity to Make Forecasts That Pass Statistical Tests

  • Chapter
Preferences and Similarities

Part of the book series: CISM International Centre for Mechanical Sciences ((CISM,volume 504))

Abstract

Defensive forecasting first identifies a betting strategy that succeeds if probabilistic forecasts are inaccurate and then makes forecasts that will defeat this strategy. Both the strategy and the forecasts are based on the similarity of the current situation to previous situations.

The theory of defensive forecasting uses the game-theoretic framework for probability, in which game theory replaces measure theory. In this framework, a classical theorem such as the law of large numbers is proven by a betting strategy that multiplies the capital it risks by a large factor if the theorem’s prediction fails. Theorems proven in this way apply not only to the classical case where only point predictions are made. Defensive forecasting is possible because the strategies are specified explicitly.

The author thanks Volodya Vovk for his advice on the exposition as well as for his central role in the work reviewed. He also thanks Hans-J. Lenz and Giacomo Della Riccia for their invitation to present this work at the 2006 ISSEK Invitational Workshop on Preferences and Similarities in Udine, Italy. Mr. Te-Chien Lo, at Rutgers Business School, helped with the final document.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution
Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD 129.00
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 169.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info
Hardcover Book
USD 169.99
Price excludes VAT (USA)
  • Durable hardcover edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Bibliography

  • José M. Bernardo and Adrian F. M. Smith. Bayesian Theory. Wiley, Chichester, 2000.

    MATH  Google Scholar 

  • L. E. J. Brouwer. Begründung der Mengenlehre unabhängig vom logischen Satz vom ausgeschlossenen Dritte. Erster Teil. Allgemeine Mengelehre. Koninklijke Nederlandse Akademie van Wetenschschappen Verhandelingen, 5:1–43, 1918.

    Google Scholar 

  • Nicolò Cesa-Bianchi and Gábor Lugosi. Prediction, Learning, and Games. Cambridge University Press, Cambridge, 2006.

    MATH  Google Scholar 

  • A. Philip Dawid. Statistical theory: The prequential approach (with discussion). Journal of the Royal Statistical Society. Series A, 147:278–292, 1984.

    Article  MATH  MathSciNet  Google Scholar 

  • A. Philip Dawid. Self-calibrating priors do not exist: Comment. Journal of the American Statistical Association, 80:340–341, 1985. This is a contribution to the discussion in Oakes 1985).

    Article  MathSciNet  Google Scholar 

  • A. Philip Dawid. Probability forecasting. In Samuel Kotz, Norman L. Johnson, and Campbell B. Read, editors, Encyclopedia of Statistical Sciences, volume 7, pages 210–218. Wiley, New York, 1986.

    Google Scholar 

  • Joseph L. Doob. Stochastic Processes. Wiley, New York, 1953.

    MATH  Google Scholar 

  • Dean P. Foster and Rakesh V. Vohra. Asymptotic calibration. Biometrika, 85:379–390, 1998.

    Article  MATH  MathSciNet  Google Scholar 

  • Peter Gács. Uniform test of algorithmic randomness over a general space. Theoretical Computer Science, 341:91–137, 2005.

    Article  MATH  MathSciNet  Google Scholar 

  • Sham Kakade and Dean Foster. Deterministic calibration and Nash equilibrium. In John Shawe-Taylor and Yoram Singer, editors, Proceedings of the Seventeenth Annual Conference on Learning Theory, volume 3120 of Lecture Notes in Computer Science, pages 33–48, Heidelberg, 2004. Springer.

    Google Scholar 

  • Andrei N. Kolmogorov. Sur la loi des grands nombres. Atti della Reale Accademia Nazionale dei Lincei, Serie VI, Rendiconti, 9:470–474, 1929.

    Google Scholar 

  • Andrei N. Kolmogorov. Sur la loi forte des grands nombres. Comptes rendus hebdomadaires des séances de l’Académie des Sciences, 191:910–912, 1930b.

    Google Scholar 

  • Masayuki Kumon and Akimichi Takemura. On a simple strategy weakly forcing the strong law of large numbers in the bounded forecasting game. Annals of the Institute of Statistical Mathematics, 2007. Forthcoming, doi: 10.1007/s10463-007-0125-5.

    Google Scholar 

  • Leonid A. Levin. Uniform tests of randomness. Soviet Mathematics Doklady, 17:337–340, 1976. The Russian original in: Доклады AH CCCP 227 (1), 1976.

    MATH  Google Scholar 

  • Leonid A. Levin. Randomness conservation inequalities; information and independence in mathematical theories Information and Control, 61: 15–37, 1984.

    Article  MATH  MathSciNet  Google Scholar 

  • Per Martin-Löf. The definition of random sequences. Information and Control, 9:602–619, 1966.

    Article  MathSciNet  Google Scholar 

  • Per Martin-Löf. Notes on Constructive Mathematics. Almqvist & Wiksell, Stockholm, 1970.

    Google Scholar 

  • James Mercer. Functions of positive and negative type, and their connection with the theory of integral equations. Philosophical Transactions of the Royal Society of London. Series A, Containing Popers of a Mathematical or Physical Character, 209:415–446, 1909.

    Article  Google Scholar 

  • David Oakes. Self-calibrating priors do not exist (with discussion). Journal of the American Statistical Association, 80:339–342, 1985.

    Article  MATH  MathSciNet  Google Scholar 

  • Herbert Robbins. Statistical methods related to the law of the iterated logarithm. Annals of Mathematical Statistics, 41: 1397–1409, 1970.

    Article  MATH  MathSciNet  Google Scholar 

  • Alvaro Sandroni, Rann Smorodinsky, and Rakesh Vohra. Calibration with many checking rules. Mathematics of Operations Research, 28:141–153, 2003.

    Article  MATH  MathSciNet  Google Scholar 

  • Bernhard Schölkopf and Alexander J. Smola. Learning with Kernels. MIT Press, Cambridge, MA, 2002.

    Google Scholar 

  • Glenn Shafer. From Cournot’s principle to the erfficient market hypothesis, November 2005. Working Paper # 15, www.probabilityandfinance.com. An abridged version appeared in Augustin Cournot: Modelling Economics, edited by J.-P. Touffut, Edward Elgar, 2007.

    Google Scholar 

  • Glenn Shafer and Vladimir Vovk. The origins and legacy of Kolmogorov’s Grundbergriffe, Working Paper 4, www.probabilityandfinance.com, October 2005. An abridged version, emphasizing the period before 1933, appeared as “The sources of Kolmogorov’s Grundbegriffe” in Statistical Science, 26:70–98, 2006.

    Google Scholar 

  • Glenn Shafer and Vladimir Vovk. Probability and Finance: It’s Only a Game. Wiley, New York, 2001.

    Book  MATH  Google Scholar 

  • Ray Solomonoff. The universal distribution and machine learning. The Computer Journal, 46, 2003.

    Google Scholar 

  • Charles J. Stone. Consistent nonparametric regression (with discussion). Annals of Statistics, 5:595–645, 1977.

    Article  MATH  MathSciNet  Google Scholar 

  • Jean-André Ville. Étude critique de la notion de collectif. Gauthier-Villars, Paris, 1939. This differs from Ville’s dissertation, which was defended in March 1939, only in that a 17-page introductory chapter replaces the dissertation’s one-page introduction. For a translation into English of the passages where Ville spells out an example of a sequence that satisfies von Mises’s conditions but can still be beat by a gambler who varies the amount he bets, see www.probabilityandfinance.com/misc/ville1939.pdf.

    Google Scholar 

  • Jean-André Ville. Notice sur les travaux scientifiques de M. Jean Ville, 1955. Prepared by Ville when he was a candidate for a position at the University of Paris and archived by the Institut Henri Poincaré in Paris.

    Google Scholar 

  • Richard von Mises. Grundlagen der Wahrscheinlichkeitsrechnung. Mathematische Zeitschrift, 5:52–99, 1919.

    Article  MathSciNet  Google Scholar 

  • Richard von Mises. Wahrscheinlichkeitsrechnung, Statistik und Wahrheit. Springer, Vienna, 1928. Second edition 1936, third 1951. A posthumous fourth edition, edited by his widow Hilda Geiringer, appeared in 1972. English editions, under the title Probability, Statistics and Truth, appeared in 1939 and 1957.

    Google Scholar 

  • Richard von Mises. Wahrscheinlichkeitsrechnung und ihre Anwendung in der Statistik und theoretischen Physik. Deuticke, Leipzig and Vienna, 1931.

    Google Scholar 

  • Vladimir Vovk. On-line regression competitive with reproducing kernel Hilbert spaces. Working Paper 11, www.probabilityandfinance.com, January 2006a.

    Google Scholar 

  • Vladimir Vovk. Non-asymptotic calibration and resolution, Working Paper 13, www.probabilityandfinance.com, July 2006b.

    Google Scholar 

  • Vladimir Vovk. Competitive on-line learning with a convex loss function, Working Paper 14, www.probabilityandfinance.com, September 2005.

    Google Scholar 

  • Vladimir Vovk. Competing with wild prediction rules, Working Paper 16, www.probabilityandfinance.com, January 2006c.

    Google Scholar 

  • Vladimir Vovk. Predictions as statements and decisions, Working Paper 17, www.probabilityandfinance.com, June 2006d.

    Google Scholar 

  • Vladimir Vovk. Leading strategies in competitive on-line prediction, Working Paper 18, www.probabilityandfinance.com, August 2007a.

    Google Scholar 

  • Vladimir Vovk. Defensive forecasting for optimal prediction with expert advice, Working Paper 20, www.probabilityandfinance.com, Augsst 2007b.

    Google Scholar 

  • Vladimir Vovk. Continuous and randomized defensive forecasting: unified view, Working Paper 21, www.probabilityandfinance.com, August 2007c.

    Google Scholar 

  • Vladimir Vovk. Derandomizing stochastic prediction strategies. Machine Learning, 35:247–282, 1999.

    Article  MATH  Google Scholar 

  • Vladimir Vovk and Glenn Shafer. The game-theoretic capital asset pricing model, Working Paper 1, www.probabilityandfinance.com, March 2002. An abridged version will appear in the International Journal of Approximate Reasoning.

    Google Scholar 

  • Vladimir Vovk and Glenn Shafer. A game-theoretic explanation of the \( \sqrt {dt} \) effect, Working Paper 5, www.probabilityandfinance.com, January 2003.

    Google Scholar 

  • Vladimir Vovk and Glenn Shafer. Good randomized sequential probability forecasting is always possible, Working Paper 7, www.probabilityaadfinance.com, September 2007. An abridged version appeared in Journal of the Royal Statistical Society, Series B, 67:747–764, 2005.

    Google Scholar 

  • Vladimir Vovk, Ilia Nouretdinov, Akimichi Takemura, and Glenn Shafer. Defensive forecasting for linear protocols, Working Paper, 10, www.probabilityandfinance.com, February 2005a. An abridged version appeared in Algorithmic Learning Theory: Proceedings of the 16th International Conference, ALT 2005, Singapore, October 8–11, 2005, edited by Sanjay Jain, Simon Ulrich Hans, and Etsuji Tomita, wwwalg.ist.hokudai.ac.jp/thomas/ALT05/alt05.jhtml on pp. 459–473 of Lecture Notes in Computer Science, Volume 3734, Springer-Verlag, 2005.

    Google Scholar 

  • Vladimir Vovk, Akimichi Takemura, and Glenn Shafer. Defensive forecasting, Working Paper 8, www.probabilityandfinance.com, January 2005b. An abridged version appeared in Proceedings of the Tenth International Workshop on Artificial Intelligence and Statistics, www.gatsby.ucl.ac.uk/aistats/.

    Google Scholar 

  • Abraham Wald. Statistical Decision Functions. Wiley, New York, 1950.

    MATH  Google Scholar 

  • Wei Wu and Glenn Shafer. Testing lead-lag effects under game-gheoretic efficient market hypothesis, Working Paper 23, www.probabilityandfinance.com, November 2007.

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Rutgers Business School and Royal Holloway, University of London, London, UK

    Glenn Shafer

Authors
  1. Glenn Shafer
    View author publications

    You can also search for this author in PubMed Google Scholar

Editor information

Editors and Affiliations

  1. University of Udine, Udine, Italy

    Giacomo Della Riccia

  2. University of Toulouse, France

    Didier Dubois

  3. University of Magdeburg, Magdeburg, Deutschland

    Rudolf Kruse

  4. Freie Universität Berlin, Berlin, Deutschland

    Hans-Joachim Lenz

Rights and permissions

Reprints and permissions

Copyright information

© 2008 CISM, Udine

About this chapter

Cite this chapter

Shafer, G. (2008). Defensive Forecasting: How to Use Similarity to Make Forecasts That Pass Statistical Tests. In: Della Riccia, G., Dubois, D., Kruse, R., Lenz, HJ. (eds) Preferences and Similarities. CISM International Centre for Mechanical Sciences, vol 504. Springer, Vienna. https://doi.org/10.1007/978-3-211-85432-7_9

Download citation

  • DOI: https://doi.org/10.1007/978-3-211-85432-7_9

  • Publisher Name: Springer, Vienna

  • Print ISBN: 978-3-211-85431-0

  • Online ISBN: 978-3-211-85432-7

  • eBook Packages: EngineeringEngineering (R0)

Publish with us

Policies and ethics

Search

Navigation