Hostname: page-component-848d4c4894-tn8tq Total loading time: 0 Render date: 2024-06-17T23:11:08.428Z Has data issue: false hasContentIssue false
  • English
  • Français

Simpson-type paradoxes, dependence, and ageing

Part of: Distribution theory - Probability Survival analysis and censored data

Published online by Cambridge University Press:  14 July 2016

Marco Scarsini  and
Fabio Spizzichino
Marco Scarsini*
Affiliation:
Università D'Annunzio
Fabio Spizzichino*
Affiliation:
Università ‘La Sapienza’
*
∗Postal address: Dipartimento di Scienze, Università D'Annunzio, Viale Pindaro 42, I-65127 Pescara, Italy. Email address: scarsini@sci.unich.it.
∗∗Postal address: Dipartimento di Matematica, Università ‘La Sapienza’, Piazzale Aldo Moro 5, I-00185 Roma, Italy. Email address: spizzichino@axrma.uniroma1.it.
Get access Rights & Permissions [Opens in a new window]

Abstract

We will state a general version of Simpson's paradox, which corresponds to the loss of some dependence properties under marginalization. We will then provide conditions under which the paradox is avoided. Finally we will relate these Simpson-type paradoxes to some well-known paradoxes concerning the loss of ageing properties when the level of information changes.

Keywords

Positive dependenceSimpson's paradoxYule's reversal paradoxweakened by failureIFRDFRMIFR

MSC classification

Primary: 62N05: Reliability and life testing
Secondary: 60E15: Inequalities; stochastic orderings 60E05: Distributions
Type
Research Papers
Information
Journal of Applied Probability , Volume 36 , Issue 1 , March 1999 , pp. 119 - 131
Copyright
Copyright © Applied Probability Trust 1999 

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

Arjas, E. (1981a). A stochastic process approach to multivariate reliability systems: Notions based on conditional stochastic order. Math. Operat. Res. 6, 263–276.Google Scholar
Arjas, E. (1981b). The failure and hazard processes in multivariate reliability theory. Math. Operat. Res. 6, 551–562.CrossRefGoogle Scholar
Arjas, E., and Norros, I. (1984). Life lengths and association: a dynamic approach. Math. Operat. Res. 9, 151–158.CrossRefGoogle Scholar
Arjas, E., and Norros, I. (1991). Stochastic order and martingale dynamics in multivariate life length models: a review. In Stochastic Orders and Decision under Risk, ed. Mosler, K. and Scarsini, M. IMS Lecture Notes/Monograph Series, Hayward, California, pp. 7–24.CrossRefGoogle Scholar
Barlow, R. E. (1985). A Bayesian explanation of an apparent failure rate paradox. IEEE Trans. Rel. 34, 107–108.Google Scholar
Barlow, R. E., and Pereira, C. A. de B. (1990). Conditional independence and probabilistic influence diagrams, in Topics in Statistical Dependence, ed. Block, H. W., Sampson, A. R. and Savits, T. H. IMS Lecture Notes/Monograph Series, Hayward, California, pp. 19–33.Google Scholar
Barlow, R. E., and Proschan, F. (1975). Statistical Theory of Reliability and Life Testing, Holt, Rinehart, and Winstion, New York, NY.Google Scholar
Blyth, C. R. (1972). On Simpson's paradox and the sure thing principle. J. Amer. Statist. Assoc. 67, 364–366.CrossRefGoogle Scholar
Blyth, C. R. (1973). Simpson's paradox and mutually favorable events. J. Amer. Statist. Assoc. 68, 746.CrossRefGoogle Scholar
Cartwright, N. (1979). Causal laws and effective strategies. Noûs 13, 419–437.Google Scholar
Cohen, J. E. (1986). An uncertainty principle in demograpy and the unisex issue. Amer. Statistician 40, 32–39.Google Scholar
Eells, E. (1987). Cartwright and Otte on Simpson's paradox. Philosophy of Science 54, 233–243.Google Scholar
Good, I. J., and Mittal, Y. (1987). The amalgamation and geometry of two-by-two contingency tables. Ann. Statist. 15, 694–711.Google Scholar
Gudder, S. P. (1988). Quantum Probability, Academic Press, Boston, MA.Google Scholar
Hardcastle, V. G. (1991). Partitions, probabilistic causal laws, and Simpson's paradox. Synthése 86, 209–228.Google Scholar
Jensen, U. (1996). Stochastic models of reliability and maintenance: an overview. In Reliability and Maintenance of Complex Systems (ed. Özekici, S.). Springer, Berlin, pp. 3–36.Google Scholar
Jogdeo, K. (1978). On a probability bound of Marshall and Olkin. Ann. Statist. 6, 232–234.Google Scholar
Karlin, S., and Rinott, Y. (1980). Classes of orderings of measures and related correlation inequalities. I. Multivariate totally positive distributions. J. Multivariate Anal. 10, 467–498.CrossRefGoogle Scholar
Kimeldorf, G., and Sampson, A. R. (1989). A framework for positive dependence. Ann. Inst. Statist. Math. 41, 31–45.Google Scholar
Langberg, N. A., Leòn, R. V., and Proschan, F. (1980). Characterization of nonparamteric classes of life distributions. Ann. Prob. 8, 1163–1170.CrossRefGoogle Scholar
Lindley, D. V., and Novik, M. R. (1981). On the role of exchangeability in inference. Ann. Statist. 9, 45–58.Google Scholar
Mittal, Y. (1991). Homogeneity of subpopulations and Simpson's paradox. J. Amer. Statist. Assoc. 86, 167–172.Google Scholar
Norros, I. (1985). Systems weakened by failures. Stochast. Proc. Appl. 20, 181–196.Google Scholar
Otte, R. (1985). Probabilistic causality and Simpson's paradox. Philosophy of Science 52, 110–125.Google Scholar
Ross, S. M. (1984). A model in which component failure rates depend on the working set. Naval Res. Logist. Quart. 31, 297–300.Google Scholar
Saari, D. G. (1995). A chaotic exploration of aggregation paradoxes. SIAM Rev. 37, 37–52.Google Scholar
Samuels, M. L. (1993). Simpson's paradox and related phenomena. J. Amer. Statist. Assoc. 88, 81–88.Google Scholar
Shaked, M., and Shanthikumar, J. G. (1988). Multivariate conditional hazard rates and the MIFRA and MIFR properties. J. Appl. Prob. 25, 150–168.CrossRefGoogle Scholar
Shaked, M., and Shanthikumar, J. G. (1990). Multivariate stochastic orderings and positive dependence in reliability theory. Math. Operat. Res. 15, 545–552.Google Scholar
Shaked, M., and Shanthikumar, J. G. (1991). Dynamic multivariate ageing. Stoch. Proc. Appl. 38, 85–97.Google Scholar
Shaked, M., and Shanthikumar, J. G. (1994). Stochastic Orders and Their Applications, Academic Press, Boston, MA.Google Scholar
Simpson, E. H. (1951). The interpretation of interaction in contingency tables. J. Roy. Statist. Soc. B 13, 238–241.Google Scholar
Vaupel, J. V., and Yashin, A. I. (1985). Some surprising effects of selection on population dynamics. Amer. Statistician 39, 176–185.Google ScholarPubMed
Yule, G. U. (1903). Notes on the theory of association of attributes in statistics. Biometrika 2, 121–134.Google Scholar