Model selection and inference : a practical information-theoretic approach

By R. Strawderman, K. Burnham, David R. Anderson. 2000

2,652 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/9836c6cfeeb7a4831062ff44c135a62a830a0c12

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

Citations in the corpus

The Portuguese Version of the Beck Depression Inventory-ii (bdi-ii): Preliminary Psychometric Data with Two Nonclinical Samples (2011, background): “Burnham and Anderson (1998) suggest that if AIC values for one model (eg, path estimates are free to be estimated) are 10 or more units lower than AIC values for a second model (eg, constrained to the same values across groups), there is strong evidence that the first model is better than the second”
Covariation between grey seal (Halichoerus grypus) abundance and natural mortality of cod (Gadus morhua) in the southern Gulf of St. Lawrence (2005, methodology): “If a particular model has 4 < ∆AICc < 7, there is evidence that the model is not the best and a value of ∆AICc > 10 is considered very strong evidence that the model is not the best (Burnham and Anderson 1998).”
No interactive effects of pesticides and plant diversity on soil microbial biomass and respiration (2009, methodology): “After fitting the full model, the respective minimum adequate model was derived using the Akaike information criterion (AIC; Burnham and Anderson, 1998).”
Presence and Absence of Bats Across Habitat Scales in the Upper Coastal Plain of South Carolina (2006, methodology): “We used Anabat II detectors (Titley Electronics, Ltd., Ballina, New South Wales, Australia) linked to laptop computers using Anabat 6.3e software to determine presence of foraging bats at 430 survey locations on the SRS from late May through July 2001.”
Predators of Greater Sage-Grouse nests identified by video monitoring (2008, methodology): “The best-approximating model ( 0, i = 0.63) was 1.7 times ( i / j) more likely to describe the data than the 0 + 1CAM model ( AICc = 1.02, i = 0.37), but 0 model did not have substantially greater support from the data (Burnham and Anderson 1998).”
Relative effects of fragment size and connectivity on bird community in the Atlantic Rain Forest: Implications for conservation (2008, methodology): “To compare the 13 models, we used the Akaike Information Criterion (AIC, Burnham and Anderson, 1998) with the small sample correction (AICc) proposed by Hurvich and Tsai (1989), generating a rank from the best to the worst model.”
Sequential settlement and site dependence in a migratory raptor (2007, methodology): “In such cases, we compared models through an information-theoretic approach and selected the one with the lowest Akaike Information Criterion (Burnham and Anderson 1998).”
Critical Demographic Parameters for Declining Songbirds Breeding in Restored Grasslands (2006, methodology): “Thus, we compared our global model (/[g 3 t]p[g 3 t]) to reduced models using AIC, adjusted for small sample sizes and overdispersion (QAICc), and Akaike weights (Burnham and Anderson 1998).”
Small-mammal density estimation: A field comparison of grid-based vs. web-based density estimators (2003, positive): “Five of the analysts used model-averaging techniques (Burnham and Anderson 1998) to calculate N values.”
COMPARATIVE SEED SHADOWS OF BIRD‐, MONKEY‐, AND WIND‐DISPERSED TREES (2005, background): “…test, the AIC does not provide a simple frequentist measure of statistical significance; models with lower AIC scores fit the data better, and a rule of thumb is that models within four AIC units of the best model are considered plausible explanations for the data (Burnham and Anderson 1998).”