A general and simple method for obtaining R2 from generalized linear mixed-effects models
Citations
2,194 citations
1,467 citations
1,389 citations
Cites background or methods from "A general and simple method for obt..."
...In the following, we present a worked example by expanding the beetle dataset that was generated for previous work [3]....
[...]
...We have reviewed methods for estimating R(2) and ICC in the past, with a particular focus on non-Gaussian response variables in the context of biological data [2,3]....
[...]
...Research papers in the field of ecology and evolution often report only regression coefficients but not variance components of GLMMs [3]....
[...]
...Furthermore, we refer to some special considerations when obtaining R(2)GLMM and ICCGLMM from binomial GLMMs for binary and proportion data, which we did not discuss in the past [2,3]....
[...]
...Each of these distributions has a theoretical variance, namely, p(2)/3, 1 and p(2)/6, respectively, which we previous referred to as distribution-specific variances [2,3] (table 2)....
[...]
1,210 citations
Cites background or methods from "A general and simple method for obt..."
...…for GLMMs is that it returns two complementary R2 values: the marginal R2 encompassing variance explained by only the fixed effects, and the conditional R2 comprising variance explained by both fixed and random effects i.e. the variance explained by the whole model (Nakagawa & Schielzeth, 2013)....
[...]
...The Nakagawa & Schielzeth (2013) R2 functions have been incorporated into several packages, including ‘MuMIn’ (Barto n, 2016) and ‘piecewiseSEM’ (Lefcheck, 2015), and Johnson (2014) has developed an extension of the functions for random slope models....
[...]
...The method that has gained the most support over recent years is that of Nakagawa & Schielzeth (2013)....
[...]
...Further reading: Nakagawa & Schielzeth (2013) provide an excellent and accessible description of the problems with, and solutions to, generalising R2 metrics to GLMMs....
[...]
...Diversemethods have been proposed to account for this in GLMMs, including multiple so-called ‘pseudo-r2’ measures of explained variance (Nagelkerke, 1991; Cox & Snell, 1989), but their performance is often unstable for mixed models and can return negative values (Nakagawa & Schielzeth, 2013)....
[...]
1,044 citations
Cites background or methods from "A general and simple method for obt..."
...The coefficient of determinationR2 is a similar statistic that quantifies the proportion of variance explained by fixed effects (marginal R2 sensu Nakagawa & Schielzeth 2013)....
[...]
...However, we have previously reviewed the equations for estimating repeatabilities and R2 from generalized linear mixed effects models (GLMMs) (Nakagawa & Schielzeth 2010, 2013)....
[...]
...Nonetheless, there are solutions to approximate repeatability for the most widely used families of generalized linearmixedmodels (GLMMs)....
[...]
...We will illustrate the features of rptR by estimating adjusted repeatabilities for Poisson data with log link for a dataset that was generated for estimating R2 in GLMMs (Nakagawa & Schielzeth 2013)....
[...]
References
38,681 citations
36,993 citations
"A general and simple method for obt..." refers background or methods in this paper
...Information criteria are used to select the ‘best’ or ‘better’ models, and they are indeed useful for selecting the most parsimonious models from a candidate model set (Burnham & Anderson 2002)....
[...]
...…provide an estimate of the relative fit of alternative models, they do not tell us anything about the absolute model fit (cf. evidence ratio; Burnham & Anderson 2002), (ii) information criteria do not provide any information on variance explained by a model (Orelien & Edwards 2008), and…...
[...]
36,760 citations
18,952 citations
18,539 citations
"A general and simple method for obt..." refers methods in this paper
...Commonly used information criteria include Akaike Information Criterion (AIC) (Akaike 1973), Bayesian information criterion (BIC), (Schwarz 1978) and the more recently proposed deviance information criterion (DIC), (Spiegelhalter et al. 2002; reviewed in Claeskens & Hjort 2009; Grueber et al. 2011; Hamaker et al. 2011)....
[...]
...Information criteria, such as Akaike Information Criterion (AIC), are usually presented as model comparison tools formixed-effectsmodels....
[...]
...Commonly used information criteria include Akaike Information Criterion (AIC) (Akaike 1973), Bayesian information criterion (BIC), (Schwarz 1978) and the more recently proposed deviance information criterion (DIC), (Spiegelhalter et al. 2002; reviewed in Claeskens & Hjort 2009; Grueber et al. 2011;…...
[...]
...Despite these limitations, when used along with other statistics such as AIC and PCV, R2GLMM will be a useful summary statistic of mixed-effects models for both biologists and other scientists alike....
[...]