Model Selection via Focused Information Criteria for Complex Data in Ecology and Evolution

Datasets encountered when examining deeper issues in ecology and evolution are often complex. This calls for careful strategies for both model building, model selection, and model averaging. Our paper aims at motivating, exhibiting, and further developing focused model selection criteria. In contexts involving precisely formulated interest parameters, these versions of FIC, the focused information criterion, typically lead to better final precision for the most salient estimates, confidence intervals, etc. as compared to estimators obtained from other selection methods. Our methods are illustrated with real case studies in ecology; one related to bird species abundance and another to the decline in body condition for the Antarctic minke whale.

[1]  Martin Jullum,et al.  Parametric or nonparametric: The FIC approach , 2018 .

[2]  T. Schweder,et al.  Phenotypic evolution studied by layered stochastic differential equations , 2012, 1301.1463.

[3]  Leo Breiman,et al.  Statistical Modeling: The Two Cultures (with comments and a rejoinder by the author) , 2001 .

[4]  Galit Shmueli,et al.  To Explain or To Predict? , 2010, 1101.0891.

[5]  B. Efron.  Estimation and Accuracy After Model Selection , 2014, Journal of the American Statistical Association.

[6]  Leo Breiman,et al.  Statistical Modeling: The Two Cultures (with comments and a rejoinder by the author) , 2001, Statistical Science.

[7]  Galit Shmueli,et al.  To Explain or To Predict? , 2010 .

[8]  Gerda Claeskens,et al.  A High-Dimensional Focused Information Criterion , 2017 .

[9]  Mark L. Taper,et al.  Model structure adequacy analysis: selecting models on the basis of their ability to answer scientific questions , 2008, Synthese.

[10]  Eugen Pircalabelu,et al.  A Focused Information Criterion for Graphical Models in fMRI Connectivity with High-Dimensional Data , 2015 .

[11]  N. Hjort,et al.  What price semiparametric Cox regression? , 2018, Lifetime Data Analysis.

[12]  Gerda Claeskens,et al.  Asymptotic post‐selection inference for the Akaike information criterion , 2018, Biometrika.

[13]  R. Tibshirani,et al.  Uniform asymptotic inference and the bootstrap after model selection , 2015, The Annals of Statistics.

[14]  Andrey L. Vasnev,et al.  The Forecast Combination Puzzle: A Simple Theoretical Explanation , 2014 .

[15]  Nils Lid Hjort,et al.  Model Selection and Model Averaging , 2001 .

[16]  N. Hjort,et al.  Statistical sightings of better angels: Analysing the distribution of battle-deaths in interstate conflict over time , 2020 .

[17]  Alan T. K. Wan,et al.  Optimal Weight Choice for Frequentist Model Average Estimators , 2011 .

[18]  N. Hjort,et al.  Recent advances in statistical methodology applied to the Hjort liver index time series (1859–2012) and associated influential factors1 , 2016 .

[19]  Nils Lid Hjort,et al.  MINIMIZING AVERAGE RISK IN REGRESSION MODELS , 2008, Econometric Theory.

[20]  Nils Lid Hjort,et al.  Fridge: Focused fine‐tuning of ridge regression for personalized predictions , 2018, Statistics in medicine.

[21]  N. Hjort,et al.  The Focused Information Criterion , 2003 .

[22]  N. Hjort,et al.  Focused information criteria for copulas , 2019, Scandinavian Journal of Statistics.

[23]  Carsten F. Dormann,et al.  Model averaging in ecology: a review of Bayesian, information-theoretic, and tactical approaches for predictive inference , 2018, Ecological Monographs.

[24]  G. Claeskens.  Focused estimation and model averaging with penalization methods: an overview , 2012 .

[25]  G. Claeskens,et al.  A High‐dimensional Focused Information Criterion , 2018 .

[26]  N. Hjort,et al.  Frequentist Model Average Estimators , 2003 .

[27]  A. Buja,et al.  Valid post-selection inference , 2013, 1306.1059.

[28]  Brian Dennis,et al.  DENSITY DEPENDENCE IN TIME SERIES OBSERVATIONS OF NATURAL POPULATIONS: ESTIMATION AND TESTING' , 1994 .