Electronic Journal of Statistics

Maximum likelihood degree of variance component models

Elizabeth Gross, Mathias Drton, and Sonja Petrović

Full-text: Open access

Abstract

Most statistical software packages implement numerical strategies for computation of maximum likelihood estimates in random effects models. Little is known, however, about the algebraic complexity of this problem. For the one-way layout with random effects and unbalanced group sizes, we give formulas for the algebraic degree of the likelihood equations as well as the equations for restricted maximum likelihood estimation. In particular, the latter approach is shown to be algebraically less complex. The formulas are obtained by studying a univariate rational equation whose solutions correspond to the solutions of the likelihood equations. Applying techniques from computational algebra, we also show that balanced two-way layouts with or without interaction have likelihood equations of degree four. Our work suggests that algebraic methods allow one to reliably find global optima of likelihood functions of linear mixed models with a small number of variance components.

Article information

Source
Electron. J. Statist., Volume 6 (2012), 993-1016.

Dates
First available in Project Euclid: 31 May 2012

Permanent link to this document
https://projecteuclid.org/euclid.ejs/1338469232

Digital Object Identifier
doi:10.1214/12-EJS702

Mathematical Reviews number (MathSciNet)
MR2988436

Zentralblatt MATH identifier
1281.62159

Subjects
Primary: 62J10: Analysis of variance and covariance
Secondary: 62F10: Point estimation

Keywords
Analysis of variance linear mixed model maximum likelihood restricted maximum likelihood variance component

Citation

Gross, Elizabeth; Drton, Mathias; Petrović, Sonja. Maximum likelihood degree of variance component models. Electron. J. Statist. 6 (2012), 993--1016. doi:10.1214/12-EJS702. https://projecteuclid.org/euclid.ejs/1338469232


Export citation

References

  • [Atk11] Quentin D. Atkinson, Phonemic diversity supports a serial founder effect model of language expansion from Africa, Science 332 (2011), no. 6027, 346–349.
  • [BHR07] Max-Louis G. Buot, Serkan Hoşten, and Donald St. P. Richards, Counting and locating the solutions of polynomial systems of maximum likelihood equations. II. The Behrens-Fisher problem, Statist. Sinica 17 (2007), no. 4, 1343–1354.
  • [CHKS06] Fabrizio Catanese, Serkan Hoşten, Amit Khetan, and Bernd Sturmfels, The maximum likelihood degree, Amer. J. Math. 128 (2006), no. 3, 671–697.
  • [DG72] Owen L. Davies and Peter L. Goldsmith (eds.), Statistical methods in research and production, 4th ed., Hafner, 1972.
  • [DM99] Eugene Demidenko and Hélène Massam, On the existence of the maximum likelihood estimate in variance components models, Sankhyā Ser. A 61 (1999), no. 3, 431–443.
  • [DSS09] Mathias Drton, Bernd Sturmfels, and Seth Sullivant, Lectures on algebraic statistics, Birkhäuser Verlag AG, Basel, Switzerland, 2009.
  • [Far06] Julian J. Faraway, Extending the linear model with R, Texts in Statistical Science Series, Chapman & Hall/CRC, Boca Raton, FL, 2006, Generalized linear, mixed effects and nonparametric regression models.
  • [HKS05] Serkan Hoşten, Amit Khetan, and Bernd Sturmfels, Solving the likelihood equations, Found. Comput. Math. 5 (2005), no. 4, 389–407.
  • [Hoc85] R. R. Hocking, The analysis of linear models, Brooks/Cole Publishing Co., Monterey, CA, 1985.
  • [HS10] Serkan Hoşten and Seth Sullivant, The algebraic complexity of maximum likelihood estimation for bivariate missing data, Algebraic and geometric methods in statistics, Cambridge Univ. Press, Cambridge, 2010, pp. 123–133.
  • [Jia07] Jiming Jiang, Linear and generalized linear mixed models and their applications, Springer Series in Statistics, Springer, New York, 2007.
  • [MN89] P. McCullagh and J. A. Nelder, Generalized linear models, 2nd ed., Monographs on Statistics and Applied Probability, Chapman & Hall, London, 1989.
  • [SCM92] Shayle R. Searle, George Casella, and Charles E. McCulloch, Variance components, Wiley Series in Probability and Mathematical Statistics: Applied Probability and Statistics, John Wiley & Sons Inc., New York, 1992, A Wiley-Interscience Publication.
  • [SM84] William H. Swallow and John F. Monahan, Monte Carlo comparison of ANOVA, MIVQUE, REML, and ML estimators of variance components, Technometrics 26 (1984), no. 1, 47–57.
  • [SO04] Hardeo Sahai and Mario Miguel Ojeda, Analysis of variance for random models. Vol. I. Balanced data, Birkhäuser Boston Inc., Boston, MA, 2004, Theory, methods, applications and data analysis.
  • [SO05] Hardeo Sahai and Mario Miguel Ojeda, Analysis of variance for random models. Vol. II. Unbalanced data, Birkhäuser Boston Inc., Boston, MA, 2005, Theory, methods, applications, and data analysis.
  • [Stu09] Bernd Sturmfels, Open problems in algebraic statistics, Emerging applications of algebraic geometry, IMA Vol. Math. Appl., vol. 149, Springer, New York, 2009, pp. 351–363.