Modeling bivariate lifetimes based on expected present values of residual lives

Original Paper


This note introduces new concepts of bivariate ageing, that are called the bivariate decreasing (increasing) expected present value residual life, and the bivariate new better (worse) than used expected present value residual life. Interpretations of these notions for the description of the ageing of systems operating under random environments, by means of time-scales and expected present values are given. They can be applied to analyze several biological and engineering systems, from the study of the joint probabilistic behaviour of their correlated components, and to assess the degree or the pattern of their ageing. We provide some examples of bivariate distributions that fulfil these properties, including these conditions for the Marshall–Olkin shock model.


Bivariate exponential Bivariate ageing Marshall–Olkin shock model Present value Joint survival function Laplace order 


  1. Al-Ruzaiza AS, El-Gohary A (2007) A new class of positively quadrant dependent bivariate distributions with Pareto. Int Math Forum 2:1259–1273Google Scholar
  2. Ali MM (1982) Distribution of linear combinations of exponential variates. Commun Stat Theory Methods 11:1453–1463CrossRefGoogle Scholar
  3. Alzaid A, Kim JS, Proschan F (1991) Laplace ordering and its applications. J Appl Probab 28:116–130CrossRefGoogle Scholar
  4. Amari SV, Misra RB (1997) Closed-form expressions for distributions of sum of exponential random variables. IEEE Trans Reliab 46:519–522CrossRefGoogle Scholar
  5. Arnold BC, Zahedi H (1988) On multivariate mean remaining life functions. J Multivar Anal 25:1–9CrossRefGoogle Scholar
  6. Balu MN, Sabnis SV (1997) Preservation of certain dependent structures under bivariate homogeneous Poisson shock models. Stat Probab Lett 35:91–100CrossRefGoogle Scholar
  7. Block HW (1977) A family of bivariate life distributions. In: Tsokos CP, Shimi IN (eds) The theory and applications of reliability, vol 1. Academic Press, NY, pp 349–371Google Scholar
  8. Block HW, Savits TH (1980) Multivariate increasing failure rate distributions. Ann Probab 8:730–801Google Scholar
  9. Buchanan HB, Singpurwalla ND (1977) Some stochastic characterization of multivariate survival. In: Tsokos CP, Shimi IN (eds) The theory and applications of reliability, vol 1. Academic Press, NY, pp 329–348Google Scholar
  10. Denuit M (2001) Laplace transform ordering of actuarial quantities. Insur Math Econ 29:83–102CrossRefGoogle Scholar
  11. El-Neweihi E, Proschan F, Sethuraman J (1983) A multivariate new better than used class derived from a shock model. Oper Res 31:177–183CrossRefGoogle Scholar
  12. Finkelstein M, Esaulova V (2005) On the weak IFR aging of bivariate lifetime distributions. Appl Stoch Models Bus Ind 21:265–272CrossRefGoogle Scholar
  13. Gasemyr J, Natvig B (1995) Using expert opinions in Bayesian prediction of component lifetimes in a shock model. Math Oper Res 20:227–242CrossRefGoogle Scholar
  14. Ghosh M, Ebrahimi N (1982) Shock models leading to increasing failure rate and decreasing mean residual life survival. J Appl Probab 19:158–166CrossRefGoogle Scholar
  15. Ghosh M, Ebrahimi N (1983) Shock models leading to multivariate NBU and NBUE distributions. In: Sen PK (ed) Contributions to statistics: essays in honour of Norman L. Johnson. North Holland, AmsterdamGoogle Scholar
  16. Hutchinson TP, Lai CD (1990) Continuous bivariate distributions, emphasizing applications. Rumsby Scientific Publishing, Adelaide, South AustraliaGoogle Scholar
  17. Johnson NL, Kotz S (1975) A vector multivariate hazard rate. J Multivar Anal 5:53–66CrossRefGoogle Scholar
  18. Karlin S (1968) Total positivity. Stanford University Press, StanfordGoogle Scholar
  19. Karlis D (2003) ML estimation for multivariate shock models via an EM algorithm. Ann Inst Stat Math 55:817–830CrossRefGoogle Scholar
  20. Klefsjö B (1980) Some Properties of the HNBUE and HNWUE Classes of Life Distributions. Research Report 1980-8, Department of Mathematical Statistics, University of UmeaGoogle Scholar
  21. Klefsjö B (1982) NBU and NBUE survival under the Marshall–Olkin Shock Model. IAPQR Trans 7:331–344Google Scholar
  22. Kordonsky KB, Gertsbakh IB (1995a) System state monitoring and lifetime scales. 1. Reliab Eng Syst Saf 47:1–14CrossRefGoogle Scholar
  23. Kordonsky KB, Gertsbakh IB (1995b) System state monitoring and lifetime scales. 2. Reliab Eng Syst Saf 49:145–354CrossRefGoogle Scholar
  24. Li H (2003) Association of multivariate phase-type distributions, with applications to shock models. Stat Probab Lett 64:1043–1059Google Scholar
  25. Li H (2008) Duality of the multivariate distributions of Marshall–Olkin type and tail dependence. Commun Stat 37:1721–1733CrossRefGoogle Scholar
  26. Li H, Xu S (2002) Directionally convex comparisons of correlated first passage times. Method Comput Appl Probab 365–379Google Scholar
  27. Lu JC (1989) Weibull extensions of the Freund and the Marshall–Olkin bivariate exponential model. IEEE Trans Reliab 38:615–619CrossRefGoogle Scholar
  28. Lu JC (1992) Bayes parameter estimation for the bivariate Weibull model of Marshall–Olkin for censored data. IEEE Trans Reliab 41:608–615CrossRefGoogle Scholar
  29. Marshall AW, Olkin I (1967) A multivariate exponential distribution. J Am Stat Assoc 62:30–44CrossRefGoogle Scholar
  30. Marshall AW, Olkin I (1979) Inequalities: theory of majorization and its applications. Academic Press, Inc, New YorkGoogle Scholar
  31. Marshall AW, Shaked M (1979) Multivariate shock models for distributions with increasing hazard average. Ann Probab 7:343–358CrossRefGoogle Scholar
  32. Meintanis SG (2007) Test of fit for Marshall–Olkin distributions with applications. J Stat Plan Inference 137:3954–3963CrossRefGoogle Scholar
  33. Nadarajah S, Kotz S (2006) Sums, products, and ratios for downton’ s bivariate exponential distribution. Stoch Environ Res Risk Assess 20:164–170CrossRefGoogle Scholar
  34. Obretenov A (1985) Some characterizations of bivariate distributions based on hazard rate properties. In: Stability problems for stochastic models, Lecture Notes in Mathematics, Springer, Berlin, pp 223–241Google Scholar
  35. Proschan F, Sullo P (1974) Estimating the parameters of a bivariate exponential distribution in several sampling situations. In: Proschan F, Serfling R (eds) Reliability and biometry. Statistical analyses of lifelengths. Society for Industrial and Applied Mathematics, Philadelphia pp 423–440Google Scholar
  36. Rolski T, Schmidli H, Schmidt V, Teugels J (1999) Stochastic processes for insurance and finance. Wiley, ChichesterCrossRefGoogle Scholar
  37. Roy D (1994) Classifications of life distributions in multivariate set up. IEEE Trans Reliab 43:224–229CrossRefGoogle Scholar
  38. Savits TH (1985) A multivariate IFR class. J Appl Probab 22:197–204CrossRefGoogle Scholar
  39. Shaked M, Shanthikumar JG (2007) Stochastic orders, Springer, BerlinGoogle Scholar
  40. Srikanth KI, Manjunath D, Manivasakan R (2002) Bivariate exponential distributions using linear structures. Sankhya Indian J Stat 64:156–166Google Scholar
  41. Yue S (2001) Applicability of the Nagao-Kadoya bivariate exponential distribution for modeling two correlated exponentially distributed variates. Stoch Environ Res Risk Assess 15:244–260CrossRefGoogle Scholar

Copyright information

© Springer-Verlag 2009

Authors and Affiliations

  1. 1.Centro de Investigación Operativa, Departamento Estadística, Matemáticas e InformáticaUniversidad Miguel HernándezOrihuelaSpain

Personalised recommendations