Model selection and multimodel inference : a practical information-theoretic approach
By K. Burnham, David R. Anderson. 2003
37,149 citations recorded by the source corpus.
The second edition of this book is unique in that it focuses on methods for making formal statistical inference from all the models in an a priori set (Multi-Model Inference). A philosophy is presented for model-based data analysis and a general strategy outlined for the analysis of empirical data. The book invites increased attention on a priori science hypotheses and modeling. Kullback-Leibler Information represents a fundamental quantity in science and is Hirotugu Akaike's basis for model selection. The maximized log-likelihood function can be bias-corrected as an estimator of expected, relative Kullback-Leibler information. This leads to Akaike's Information Criterion (AIC) and various extensions. These methods are relatively simple and easy to use in practice, but based on deep statistical theory. The information theoretic approaches provide a unified and rigorous theory, an extension of likelihood theory, an important application of information theory, and are objective and practical to employ across a very wide class of empirical problems. The book presents several new ways to incorporate model selection uncertainty into parameter estimates and estimates of precision. An array of challenging examples is given to illustrate various technical issues. This is an applied book written primarily for biologists and statisticians wanting to make inferences from multiple models and is suitable as a graduate text or as a reference for professional analysts.
Subfields: Artificial Intelligence, Statistics and Probability
View Paper: https://www.semanticscholar.org/paper/59bb17b27d7220e930ee6bebe1f94cf43ad42c2d
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- H. Schielzeth
- Yongfeng Zhang
- Shinichi Nakagawa
- Aki Vehtari
- A. C. Davison
- Gerda Claeskens
- S. Richards
- Bruce N. McLellan
In Statistics and Probability
- Xinyu Zhang
- Hua Liang
- Shinichi Nakagawa
- Mark F. J. Steel
- Gerda Claeskens
- J. R. Magnus
- Guohua Zou
- B. Hansen
Citations in the corpus
- Paintings predict the distribution of species, or the challenge of selecting environmental predictors and evaluation statistics (2018, methodology): “In this regard, our results showed that, as for AUC, c. 30% of models based on pseudopredictors were better evaluated (lower AICc) than those based on real environment variables.”
- Response of arthropod species richness and functional groups to urban habitat structure and management (2010, methodology): “theoretic approach (Burnham and Anderson 2002; Johnson and Omland 2004) in which a-priori models were ranked according to their support by the data using Akaike weights obtained from the Akaike’s Information Criterion corrected for small sample size (AICc).”
- ESTIMATING GRIZZLY BEAR DISTRIBUTION AND ABUNDANCE RELATIVE TO HABITAT AND HUMAN INFLUENCE (2004, methodology): “We used information–theoretic methods (Burnham and Anderson 2002) with Akaike’s Information Criterion (AIC; Akaike 1973) to rank all possible parameter subset models in terms of parsimony and prediction.”
- Do tigers displace leopards? If so, why? (2010, methodology): “grassland/successional forest), total density of wild ungulates (TT), density of medium-sized ungulates (MD: axis deer, hog deer, wild boar and muntjac) and density of large ungulates (LG: barasingha + nilgai) R(2) is an adjusted coefficient of determination for logistic regression (Liao and McGee 2”
- Measuring Effective Teacher-Student Interactions From a Student Perspective (2015, methodology): “The AIC was lower by 108.453 in the three-factor model, which indicates that the one-factor model is a likely worse fit than the three-factor model (Burnham & Anderson, 2002).”
- Model selection procedures in social research: Monte-Carlo simulation results (2008, methodology): “Burnham and Anderson [2] ‘question (deny, actually) the concept of a simple ‘true model’ in the biological”
- Evaluation of consensus methods in predictive species distribution modelling (2009, methodology): “This approach is, to some extent, analogous to the method described by Burnham & Anderson (2002).”
- Changing philosophies and tools for statistical inferences in behavioral ecology (2009, methodology): “Behavioral Ecology doi:10.1093/beheco/arp137 Advance Access publication 2 November 2009 Review Changing philosophies and tools for statistical inferences in behavioral ecology”
- Changes in coexistence mechanisms along a long‐term soil chronosequence revealed by functional trait diversity (2012, methodology): “In all analyses, evidence for nonlinear (logarithmic, quadratic and power) relationships was assessed using Akaike Information Criterion (AIC) weights (Burnham & Anderson 2002).”