MuMIn (version 1.47.5)

MuMIn-package: Multi-model inference


The package MuMIn contains functions to streamline information-theoretic model selection and carry out model averaging based on information criteria.



Kamil Bartoń


The suite of functions includes:


performs automated model selection by generating subsets of the supplied ‘global’ model and optional choices of other model properties (such as different link functions). The set of models can be generated with ‘all possible’ combinations or tailored according to specified conditions.


creates a model selection table from selected models.


calculates model-averaged parameters, along with standard errors and confidence intervals. The predict method produces model-averaged predictions.


calculates the second-order Akaike information criterion. Some other criteria are provided, see below.

stdize, stdizeFit, std.coef,

can be used to standardise data and model coefficients by standard deviation or partial standard deviation.

For a complete list of functions, use library(help = "MuMIn").

By default, AIC\(_{c}\) is used to rank models and obtain model weights, although any information criterion can be used. At least the following are currently implemented in R: AIC and BIC in package stats, and QAIC, QAICc, ICOMP, CAICF, and Mallows' Cp in MuMIn. There is also a DIC extractor for MCMC models and a QIC for GEE.

Many common modelling functions in R are supported. For a complete list, see the list of supported models.

In addition to “regular” information criteria, model averaging can be performed using various types of model weighting algorithms: Bates-Granger, bootstrapped, cos-squared, jackknife, stacking, or ARM. These weighting functions are mainly applicable to glms.


Burnham, K. P. and Anderson, D. R. 2002 Model selection and multimodel inference: a practical information-theoretic approach. 2nd ed. New York, Springer-Verlag.

See Also

AIC, step or stepAIC for stepwise model selection by AIC.


Run this code
oop <- 
options(na.action = "") #  change the default "na.omit" to prevent models 
                               #  from being fitted to different datasets in 
                               #  case of missing values.

fm1 <- lm(y ~ ., data = Cement)
ms1 <- dredge(fm1)

# Visualize the model selection table:
 if(require(graphics)) { 
par(mar = c(3,5,6,4))
plot(ms1, labAsExpr = TRUE)
model.avg(ms1, subset = delta < 4)

confset.95p <- get.models(ms1, cumsum(weight) <= .95)
avgmod.95p <- model.avg(confset.95p)

Run the code above in your browser using DataLab