Statistical Papers

, Volume 54, Issue 1, pp 47–69 | Cite as

Alternative estimator for the parameters of a mixture of two binomial distributions

  • Roberto da Costa Quinino
  • Linda Lee Ho
  • Emílio Suyama
Regular Article

Abstract

The method of moments has been widely used as a simple alternative to the maximum likelihood method, mainly because of its efficiency and simplicity in obtaining parameter estimators of a mixture of two binomial distributions. In this paper, an alternative estimate is proposed which is as competitive as of the method of moments when comparing the mean squared error and computational effort.

Keywords

Mixture of two binomial distributions Method of moments Maximum likelihood Simple majority results Repetitive classifications Misclassifications Monte Carlo simulation 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Blischke WR (1962) Moment estimators for the parameters of a mixture of two binomial distributions. Anna Math Stat 33: 444–454MathSciNetMATHCrossRefGoogle Scholar
  2. Blischke WR (1964) Estimating the parameters of mixtures of binomial distributions. J Am Stat Assoc 59: 510–528MathSciNetMATHCrossRefGoogle Scholar
  3. Bross I (1954) Misclassification in 2 × 2 tables. Biometrics 10: 478–486MathSciNetMATHCrossRefGoogle Scholar
  4. Burke RJ, Davis RD, Kaminsky FC, Roberts AEP (1995) The effect of inspector errors on the true fraction non-conforming: an industrial experiment. Qual Eng 7(3): 543–550CrossRefGoogle Scholar
  5. Everitt BS, Hand DJ (1981) Finite mixture distributions. Chapman and Hall, LondonMATHCrossRefGoogle Scholar
  6. Fujisawa H., Izume S (2000) Inference about misclassification probabilities from repeated binary responses. Biometrics 56: 706–711MATHCrossRefGoogle Scholar
  7. Greenberg BS, Stokes SL (1995) Repetitive testing in the presence of inspection errors. Technometrics 37(1): 102–111MATHCrossRefGoogle Scholar
  8. Gustafson P (2003) Measurement error and misclassification in statistics and epidemiology: impacts and bayesian adjustments. Chapman & Hall, New YorkCrossRefGoogle Scholar
  9. Johnson NL, Kotz S (1985) Some tests for detection of faulty inspection. Stat Papers 26: 19–29MathSciNetGoogle Scholar
  10. Johnson NL, Kotz S (1988) Estimation from binomial data with classifiers of known and unknown imperfections. Naval Res Logist 35: 147–156MathSciNetMATHCrossRefGoogle Scholar
  11. Johnson NL, Kotz S, Kemp AW (1992) Univariete discrete distributions. John Wiley & Sons, Inc, New YorkGoogle Scholar
  12. Johnson NL, Kotz S, Wu X (1991) Inspection errors for attributes in quality control. Chapman & Hall, LondonGoogle Scholar
  13. Kukush A, Schneeweis H, Wolf R (2004) Three estimators for the Poisson regression model with measurement errors. Stat Papers 45: 351–368MathSciNetMATHCrossRefGoogle Scholar
  14. Lindsey BG (1985) Errors in inspection: integer parameter maximum likelihood in a finite population. J Am Stat Assoc 392(80): 879–885CrossRefGoogle Scholar
  15. McLachlan G, Peel D (2000) Finite mixture models. John Wiley & Sons, Inc, New YorkMATHCrossRefGoogle Scholar
  16. Montgomery DC (2001) Introduction to statistical quality control. 4 ed. John Wiley, New YorkGoogle Scholar
  17. Pepe MSP, Janes H (2007) Insights into latent class analysis of diagnostic test performance. Biostatistics 8(2): 474–484MATHCrossRefGoogle Scholar
  18. Pham-Gia T, Turkkan N (2005) Bayesian decision criteria in the presence of noises under quadratic and absolute value loss functions. Stat Papers 46: 247–266MathSciNetMATHCrossRefGoogle Scholar
  19. Shisong M, Kotz S, Johnson NL (1986) Estimation by moments in a model of faulty inspection. J Stat Comput Simul 23(4): 265–272MATHCrossRefGoogle Scholar
  20. Teicher H (1961) Identifiability of mixtures. Ann Math Stat 32: 244–248MathSciNetMATHCrossRefGoogle Scholar
  21. Van Wieringen WN (2005) On identifiability of certain latent class models. Stat Probab Lett 75: 211–218MathSciNetMATHCrossRefGoogle Scholar
  22. Van Wieringen WN, Mast J (2008) Measurement system analysis for binary data. Technometrics 50(4): 468–478MathSciNetCrossRefGoogle Scholar

Copyright information

© Springer-Verlag 2011

Authors and Affiliations

  • Roberto da Costa Quinino
    • 1
  • Linda Lee Ho
    • 2
  • Emílio Suyama
    • 1
  1. 1.Depto EstatísticaICEx, UFMGBelo HorizonteBrazil
  2. 2.Depto Engenharia de ProduçãoEPUSPSão PauloBrazil

Personalised recommendations