AIC model selection and multimodel inference in behavioral ecology: some background, observations, and comparisons

An Erratum to this article is available

Abstract

We briefly outline the information-theoretic (I-T) approaches to valid inference including a review of some simple methods for making formal inference from all the hypotheses in the model set (multimodel inference). The I-T approaches can replace the usual t tests and ANOVA tables that are so inferentially limited, but still commonly used. The I-T methods are easy to compute and understand and provide formal measures of the strength of evidence for both the null and alternative hypotheses, given the data. We give an example to highlight the importance of deriving alternative hypotheses and representing these as probability models. Fifteen technical issues are addressed to clarify various points that have appeared incorrectly in the recent literature. We offer several remarks regarding the future of empirical science and data analysis under an I-T framework.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3

References

  1. Akaike H (1973) Information theory as an extension of the maximum likelihood principle. In: Petrov BN, Csaki F (eds) Second international symposium on information theory. Akademiai Kiado, Budapest, pp 267–281

    Google Scholar 

  2. Akaike H (1974) A new look at the statistical model identification. IEEE Trans Automat Contr AC 19:716–723

    Article  Google Scholar 

  3. Akaike H (1977) On entropy maximization principle. In: Krishnaiah PR (ed) Applications of statistics. North-Holland, Amsterdam, pp 27–41

    Google Scholar 

  4. Akaike H (1978a) On the likelihood of a time series model. Statistician 27:217–235

    Article  Google Scholar 

  5. Akaike H (1978b) Bayesian analysis of the minimum AIC procedure. Ann Inst Stat Math, Part A 30:9–14

    Article  Google Scholar 

  6. Akaike H (1979) A Bayesian extension of the minimum AIC procedure of autoregressive model fitting. Biometrika 66:237–242

    Article  Google Scholar 

  7. Akaike H (1981a) Likelihood of a model and information criteria. J Econom 16:3–14

    Article  Google Scholar 

  8. Akaike H (1981b) Modern development of statistical methods. In: Eykhoff P (ed) Trends and progress in system identification. Pergamon, Paris, pp 169–184

    Google Scholar 

  9. Akaike H (1983a) Statistical inference and measurement of entropy. In: Box GEP, Leonard T, Wu C-F (eds) Scientific inference, data analysis, and robustness. Academic, London, pp 165–189

    Google Scholar 

  10. Akaike H (1983b) Information measures and model selection. Int Statistical Institute 44:277–291

    Google Scholar 

  11. Akaike H (1985) Prediction and entropy. In: Atkinson AC, Fienberg SE (eds) A celebration of statistics. Springer, New York, pp 1–24

    Google Scholar 

  12. Akaike H (1992) Information theory and an extension of the maximum likelihood principle. In: Kotz S, Johnson NL (eds) Breakthroughs in statistics, vol 1. Springer, London, pp 610–624

    Google Scholar 

  13. Akaike H (1994) Implications of the informational point of view on the development of statistical science. In: Bozdogan H (ed) Engineering and scientific applications. Vol. 3, Proceedings of the first US/Japan conference on the frontiers of statistical modeling: an informational approach. Kluwer, Dordrecht, pp 27–38

    Google Scholar 

  14. Anderson DR (2008) Model based inference in the life sciences: a primer on evidence. Springer, New York

    Google Scholar 

  15. Anderson DR, Burnham KP, Thompson WL (2000) Null hypothesis testing: problems, prevalence, and an alternative. J Wildl Manage 64:912–923

    Article  Google Scholar 

  16. Azzalini A (1996) Statistical inference based on the likelihood. Chapman and Hall, London

    Google Scholar 

  17. Brylawski AMZ, Whittingham LA (2004) An experimental study of mate guarding and paternity in house wrens. Anim Behav 68:1417–1424

    Article  Google Scholar 

  18. Burnham KP, Anderson DR (1992) Data-based selection of an appropriate biological model: the key to modern data analysis. In: McCullough DR, Barrett RH (eds) Wildlife 2001: populations. Elsevier, London, pp 16–30

    Google Scholar 

  19. Burnham KP, Anderson DR (2001) Kullback-Leibler information as a basis for strong inference in ecological studies. Wildlife Research 28:111–119

    Article  Google Scholar 

  20. Burnham KP, Anderson DR (2002) Model selection and multimodel inference: a practical information-theoretic approach, 2nd edn. Springer, New York

    Google Scholar 

  21. Burnham KP, Anderson DR (2004) Multimodel inference: understanding AIC and BIC in model selection. Sociol Methods Res 33:261–304

    Article  Google Scholar 

  22. Chamberlin TC (1890) The method of multiple working hypotheses. Science 15:92–96, Reprinted 1965, Science 148:754-759

    Article  Google Scholar 

  23. Cover TM, Thomas JA (2006) Elements of information theory, 2nd edn. Wiley, New York

    Google Scholar 

  24. Currie D, Krupa AP, Burke T, Thompson DBA (1999) The effect of experimental male removals on extrapair paternity in the wheatear, Oenanthe oenanthe. Anim Behav 57:145–152

    PubMed  Article  Google Scholar 

  25. Dayton CM (1988) Information criteria for the paired-comparisons problem. Am Stat 52:144–151

    Article  Google Scholar 

  26. Dochtermann N, Jenkins SH (2010) Developing and evaluating candidate hypotheses in behavioral ecology. Behav Ecol Sociobiol doi:10.1007/s00265-010-1039-4

  27. Edwards AWF (1992) Likelihood: expanded edition. Johns Hopkins University Press, Baltimore

    Google Scholar 

  28. Eimes JA, Parker PG, Brown JL, Brown ER (2005) Extrapair fertilization and genetic similarity of social mates in the Mexican jay. Behav Ecol 16:456–460

    Article  Google Scholar 

  29. Elliott LP, Brook BW (2007) Revisiting Chamberlin (1890): multiple working hypotheses for the 21st century. Bioscience 57:608–614

    Article  Google Scholar 

  30. Findley DF, Parzen E (1995) A conversation with Hirotugu Akaike. Stat Sci 10:104–117

    Article  Google Scholar 

  31. Freckleton RP (2010) Dealing with collinearity in behavioural and ecological data: model averaging and the problems of measurement error. Behav Ecol Sociobiol doi:10.1007/s00265-010-1045-6

  32. Garamszegi LZ (2010) Information-theoretic approaches to statistical analysis in behavioral ecology: an introduction. Behav Ecol Sociobiol doi:10.1007/s00265-010-1028-7

  33. Griffith SC, Owens IPF, Thuman KA (2002) Extra-pair paternity in birds: a review of interspecific variation and adaptive function. Mol Ecol 11:2195–2212

    PubMed  Article  CAS  Google Scholar 

  34. Guiasu S (1977) Information theory with applications. McGraw Hill, New York

    Google Scholar 

  35. Hall ML (2004) A review of hypotheses for the functions of avian duetting. Behav Ecol Sociobiol 55:415–430

    Article  Google Scholar 

  36. Hobson A, Cheng B-K (1973) A comparison of the Shannon and Kullback information measures. J Stat Phys 7:301–310

    Article  Google Scholar 

  37. Hoeting JA, Madigan D, Raftery AE, Volinsky CT (1999) Bayesian model averaging: a tutorial (with discussion). Stat Sci 14:382–417

    Article  Google Scholar 

  38. Hoi-Leitner M, Hoi H, Romero-Pujante M, Valera F (1999) Female extra-pair behaviour and environmental quality in the serin (Serinus serinus): a test of the ‘Constrained Female Hypothesis’. Proc Biol Sci 266:1021–1026

    Article  Google Scholar 

  39. Hurvich CM, Tsai C-L (1989) Regression and time series model selection in small samples. Biometrika 76:297–307

    Article  Google Scholar 

  40. Jessop A (1995) Informed assessments: an introduction to information, entropy and statistics. Ellis Horwood, London

    Google Scholar 

  41. Konishi S, Kitagawa G (2007) Information criteria and statistical modeling. Springer, New York

    Google Scholar 

  42. Kullback S, Leibler RA (1951) On information and sufficiency. Ann Math Stat 22:79–86

    Article  Google Scholar 

  43. Lack D (1968) Ecological adaptations for breeding in birds. Methuen and Company, London

    Google Scholar 

  44. McCullagh P, Nelder JA (1989) Generalized linear models, 2nd edn. Chapman and Hall/CRC, Boca Raton

    Google Scholar 

  45. Mennill DJ, Ramsay SM, Boag PT, Ratcliffe LM (2004) Patterns of extrapair mating in relation to male dominance status and female nest placement in black-capped chickadees. Behav Ecol 15:757–765

    Article  Google Scholar 

  46. Mundry R, Numm CL (2009) Stepwise model fitting and statistical inference: turning noise into signal pollution. Am Nat 173:119–123

    PubMed  Article  Google Scholar 

  47. Neudorf DLH (2004) Extrapair paternity in birds: understanding variation among species. Auk 121:302–307

    Article  Google Scholar 

  48. Pan W (2001a) Akaike’s information criterion in generalized estimating equations. Biometrics 57:120–125

    PubMed  Article  CAS  Google Scholar 

  49. Pan W (2001b) Model selection in estimating equations. Biometrics 57:529–534

    PubMed  Article  CAS  Google Scholar 

  50. Parzen E (1994) Hirotugu Akaike, statistical scientist. In: Bozdogan H (ed) Engineering and Scientific Applications, vol. 1. Proceedings of the First US/Japan Conference on the Frontiers of Statistical Modeling: An Informational Approach. Kluwer, Dordrecht, Netherlands. pp. 25–32

  51. Parzen E, Tanabe K, Kitagawa G (1998) Selected papers of Hirotugu Akaike. Springer, New York

    Google Scholar 

  52. Pawitan Y (2001) In all likelihood: statistical modelling and inference using likelihood. Oxford Science, Oxford

    Google Scholar 

  53. Perreault S, Lemon RE, Kuhnlein U (1997) Patterns and correlates of extrapair paternity in American redstarts (Setophaga ruticilla). Behav Ecol 8:612–621

    Article  Google Scholar 

  54. Platt JR (1964) Strong inference. Science 146:347–353

    PubMed  Article  CAS  Google Scholar 

  55. Qin J, Lawless G (1994) Empirical likelihood and general estimating equations. Ann Stat 22:300–325

    Article  Google Scholar 

  56. Richardson DS, Burke T (2001) Extrapair paternity and variance in reproductive success related to breeding density in Bullock’s orioles. Anim Behav 62:519–525

    Article  Google Scholar 

  57. Rissanen J (2007) Information and complexity in statistical modeling. Springer, New York

    Google Scholar 

  58. Royall RM (1997) Statistical evidence: a likelihood paradigm. Chapman and Hall, London

    Google Scholar 

  59. Rubinstein DR (2007) Territory quality drives intraspecific patterns of extrapair paternity. Behav Ecol 18:1058–1064

    Article  Google Scholar 

  60. Schmoll T, Quellmalz A, Dietrich V, Winkel W, Epplen JT, Lubjuhn T (2005) Genetic similarity between pair mates is not related to extrapair paternity in the socially monogamous coal tit. Anim Behav 69:1013–1022

    Article  Google Scholar 

  61. Schmoll T, Mund V, Dietrich-Bischoff V, Winkel W, Lubjuhn T (2007) Male age predicts extrapair and total fertilization success in the socially monogamous coal tit. Behav Ecol 18:1073–1081

    Article  Google Scholar 

  62. Schwarz G (1978) Estimating the dimension of a model. Ann Stat 6:461–464

    Article  Google Scholar 

  63. Sellke T, Bayarri MJ, Berger JO (2001) Calibration of p values for testing precise null hypotheses. Am Stat 55:62–71

    Article  Google Scholar 

  64. Severini TA (2000) Likelihood methods in statistics. Oxford University Press, Oxford

    Google Scholar 

  65. Shannon CE (1948) A mathematical theory of communication. Bell Syst Tech J 27:379–423, and 27:623–656

    Google Scholar 

  66. Soofi ES (1994) Capturing the intangible concept of information. J Am Stat Assoc 89:1243–1254

    Article  Google Scholar 

  67. Steidl RJ (2007) Model selection, hypothesis testing, and risks of condemning analytical tools. J Wildl Manage 70:1497–1498

    Article  Google Scholar 

  68. Sugiura N (1978) Further analysis of the data by Akaike’s information criterion and the finite corrections. Commun Stat, Theory Methods A7:13–26

    Google Scholar 

  69. Symonds M, Moussalli A (2010) Model selection, multimodel inference and model averaging using Akaike’s information criterion: an introduction for statistically terrified behavioural ecologists. Behavioral Ecology and Sociobiology doi:10.1007/s00265-010-1037-6

  70. Westneat DF, Stewart IRK (2003) Extra-pair paternity in birds: causes, correlates and conflict. Ann Rev Ecolog Syst 34:365–396

    Article  Google Scholar 

  71. Whittingham MJ, Stephens PA, Bradbury RB, Freckleton RP (2006) Why do we still use stepwise modeling in ecology and behaviour? J Anim Ecol 75:1182–1189

    PubMed  Article  Google Scholar 

  72. Yezerinac SM, Weatherhead PJ (1997) Extra-pair mating, male plumage coloration and sexual selection in yellow warblers (Dendroica petechia). Proc R Soc Lond B 264:527–532

    Article  Google Scholar 

Download references

Acknowledgments

The authors thank The Colorado Cooperative Fish and Wildlife Research Unit and the Department of Fish, Wildlife, and Conservation Biology at Colorado State University for continuous support. We appreciate the help of Robert Montgomerie and that of two anonymous reviewers as these helped improve the manuscript.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Kenneth P. Burnham.

Additional information

This contribution is part of the Special Issue “Model selection, multimodel inference and information-theoretic approaches in behavioral ecology” (see Garamszegi 2010).

An erratum to this article can be found at http://dx.doi.org/10.1007/s00265-010-1084-z

Communicated by L. Garamszegi

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Burnham, K.P., Anderson, D.R. & Huyvaert, K.P. AIC model selection and multimodel inference in behavioral ecology: some background, observations, and comparisons. Behav Ecol Sociobiol 65, 23–35 (2011). https://doi.org/10.1007/s00265-010-1029-6

Download citation

Keywords

  • AIC
  • Evidence
  • Kullback–Leibler information
  • Model averaging
  • Model likelihoods
  • Model probabilities
  • Model selection
  • Multimodel inference