Statistical Papers

, Volume 53, Issue 4, pp 1001–1014 | Cite as

A fast robust method for fitting gamma distributions

  • Brenton R. Clarke
  • Peter L. McKinnon
  • Geoff Riley
Regular Article


The art of fitting gamma distributions robustly is described. In particular we compare methods of fitting via minimizing a Cramér Von Mises distance, an L 2 minimum distance estimator, and fitting a B-optimal M-estimator. After a brief prelude on robust estimation explaining the merits in terms of weak continuity and Fréchet differentiability of all the aforesaid estimators from an asymptotic point of view, a comparison is drawn with classical estimation and fitting. In summary, we give a practical example where minimizing a Cramér Von Mises distance is both efficacious in terms of efficiency and robustness as well as being easily implemented. Here gamma distributions arise naturally for “in control” representation indicators from measurements of spectra when using fourier transform infrared (FTIR) spectroscopy. However, estimating the in-control parameters for these distributions is often difficult, due to the occasional occurrence of outliers.


Gamma distributions Fréchet differentiability Weak continuity Robust estimation Minimum distance estimation 

Mathematics Subject Classification (2000)

62F35 62-07 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Ahmed AN, Abouammoh AM (1993) Characterizations of gamma, inverse Gaussian, and negative binomial distributions via their length-biased distributions. Stat Papers 34: 167–173MathSciNetzbMATHCrossRefGoogle Scholar
  2. Andrews DF, Bickel PJ, Hampel FR, Huber PJ, Rogers WH, Tukey JW (1972) Robust estimates of location: advances. Princeton University Press,Google Scholar
  3. Bednarski T, Clarke BR (1998) On locally uniform expansions of regular functionals. Discuss Math Algebr Stoch Methods 18: 155–165MathSciNetzbMATHGoogle Scholar
  4. Bednarski T, Clarke BR, Kolkiewicz W (1991) Statistical expansions and locally uniform Fréchet differentiability. J Aust Math Soc A 50: 88–97MathSciNetzbMATHCrossRefGoogle Scholar
  5. Boos DD (1981) Minimum distance estimators for location and goodness of fit. J Am Stat Assoc 76: 663–670MathSciNetzbMATHCrossRefGoogle Scholar
  6. Bowman KO, Shenton LR (1988) Properties of estimators for the gamma distribution. Marcel Dekker Inc, New YorkzbMATHGoogle Scholar
  7. Clarke BR (1983) Uniqueness and Fréchet differentiability of functional solutions to maximum likelihood equations. Ann Stat 11: 1196–1205zbMATHGoogle Scholar
  8. Clarke BR (1986) Nonsmooth analysis and Fréchet differentiability of M-functionals. Probab Theory Relat Fields 73: 197–209MathSciNetzbMATHCrossRefGoogle Scholar
  9. Clarke BR (1989) An unbiased minimum distance estimator of the proportion parameter in a mixture of two normal distributions. Stat Probab Lett 7: 275–281zbMATHCrossRefGoogle Scholar
  10. Clarke BR (2000a) A remark on robustness and weak continuity of M-estimators. J Aust Math Soc A 68: 411–418zbMATHCrossRefGoogle Scholar
  11. Clarke BR (2000b) A review of differentiability in relation to robustness with an application to seismic data analysis. Proc Indian Natl Sci Acad A 66: 467–482Google Scholar
  12. Clarke BR, Heathcote CR (1994) Robust estimation of k-component univariate normal mixtures. Ann Inst Stat Math 46: 83–93MathSciNetzbMATHCrossRefGoogle Scholar
  13. Clarke BR, McKinnon PL (2005) Robust inference and modeling for the single ion channel. J Stat Comput Simul 75: 513–529MathSciNetzbMATHCrossRefGoogle Scholar
  14. Donoho DL, Liu RC (1988) The “automatic” robustness of minimum distance functionals. Ann Stat 16: 552–586MathSciNetzbMATHCrossRefGoogle Scholar
  15. Eyer S, Riley G (1999) Measurement quality assurance in a production system for bauxite analysis by FTIR. N Am Chapter Int Chemom Soc. Newsl No. 19Google Scholar
  16. Hampel FR, Ronchetti EM, Rousseeuw PJ, Stahel WJ (1986) Robust statistics, the approach based on influence functions. Wiley, New YorkzbMATHGoogle Scholar
  17. Heathcote CR, Silvapulle PMJ (1981) Minimum mean squared estimation of location and scale parameters under misspecification of the model. Biometrika 64: 501–504MathSciNetCrossRefGoogle Scholar
  18. Hettmansperger TP, Hueter I, Hüsler J (1994) Minimum distance estimators. J Stat Plan Inference 41: 291–302zbMATHCrossRefGoogle Scholar
  19. Jackson JE (1991) A user’s guide to principal components. Wiley, New YorkzbMATHCrossRefGoogle Scholar
  20. Kass RE, Steffey D (1989) Approximate Bayesian inference in conditionally independent hierarchical models. J Am Stat Assoc 84: 717–726MathSciNetCrossRefGoogle Scholar
  21. Keating JP, Glaser RE, Ketchum NS (1990) Testing hypotheses about the shape parameter of a gamma distribution. Technometrics 32: 67–82MathSciNetzbMATHCrossRefGoogle Scholar
  22. Marazzi A, Ruffieux C (1996) Implementing M-estimators of the gamma distribution. In: Rieder H (ed) Robust statistics, data analysis, and computer intensive methods. In: Honor of Peter Huber’s 60th birthday. Lecture notes in statistics. Springer, Heidelberg, pp 277–297Google Scholar
  23. Marazzi A, Ruffieux C (1999) The truncated mean of an asymetric distribution. Comp Stat Data Anal 32: 79–100MathSciNetCrossRefGoogle Scholar
  24. Parr WC (1985) Jacknifing differentiable statistical functionals. J R Stat Soc B 47: 56–66MathSciNetzbMATHGoogle Scholar
  25. Wong A (1995) On approximate inference for the two-parameter gamma model. Stat Papers 36: 49–59zbMATHCrossRefGoogle Scholar
  26. Woodward WA, Parr WC, Schucany WR, Lindsey H (1984) A comparison of minimum distance and maximum likelihood estimation of a mixture proportion. J Am Stat Assoc 79: 590–598MathSciNetzbMATHCrossRefGoogle Scholar

Copyright information

© Springer-Verlag 2011

Authors and Affiliations

  • Brenton R. Clarke
    • 1
  • Peter L. McKinnon
    • 2
  • Geoff Riley
    • 3
  1. 1.Mathematics and Statistics, School of Chemical and Mathematical SciencesMurdoch UniversityMurdochAustralia
  2. 2.Mathematics and StatisticsCurtin UniversityBentleyAustralia
  3. 3.Technology Delivery GroupAlcoa World AluminaKwinanaAustralia

Personalised recommendations