Select a formula-based model by AIC.
step(object, scope, scale = 0,
direction = c("both", "backward", "forward"),
trace = 1, keep = NULL, steps = 1000, k = 2, …)
an object representing a model of an appropriate class (mainly
This is used as the initial model in the stepwise search.
defines the range of models examined in the stepwise search.
This should be either a single formula, or a list containing
lower, both formulae. See the
details for how to specify the formulae and how they are used.
the mode of stepwise search, can be one of
"forward", with a default of
scope argument is missing the default for
"backward". Values can be abbreviated.
if positive, information is printed during the running of
Larger values may give more detailed information.
a filter function whose input is a fitted model object and the
AIC statistic, and whose output is arbitrary.
keep will select a subset of the components of
the object and return them. The default is not to keep anything.
the maximum number of steps to be considered. The default is 1000 (essentially as many as required). It is typically used to stop the process early.
the multiple of the number of degrees of freedom used for the penalty.
k = 2 gives the genuine AIC:
k = log(n) is sometimes
referred to as BIC or SBC.
any additional arguments to
the stepwise-selected model is returned, with up to two additional
components. There is an
"anova" component corresponding to the
steps taken in the search, as well as a
"keep" component if the
keep= argument was supplied in the call. The
"Resid. Dev" column of the analysis of deviance table refers
to a constant minus twice the maximized log likelihood: it will be a
deviance only in cases where a saturated model is well-defined
The model fitting must apply the models to the same dataset. This
may be a problem if there are missing values and R's default of
na.action = na.omit is used. We suggest you remove the
missing values first.
Calls to the function
nobs are used to check that the
number of observations involved in the fitting process remains unchanged.
repeatedly; it will work for any method for which they work, and that
is determined by having a valid method for
When the additive constant can be chosen so that AIC is equal to
Mallows' \(C_p\), this is done and the tables are labelled
The set of models searched is determined by the
The right-hand-side of its
lower component is always included
in the model, and right-hand-side of the model is included in the
upper component. If
scope is a single formula, it
upper component, and the
lower model is
scope is missing, the initial model is used as the
Models specified by
scope can be templates to update
object as used by
update.formula. So using
. in a
scope formula means ‘what is
already there’, with
.^2 indicating all interactions of
There is a potential problem in using
glm fits with a
scale, as in that case the deviance is not simply
related to the maximized log-likelihood. The
"glm" method for
extractAIC makes the
appropriate adjustment for a
gaussian family, but may need to be
amended for other cases. (The
families have fixed
scale by default and do not correspond
to a particular maximum-likelihood problem for variable
Hastie, T. J. and Pregibon, D. (1992) Generalized linear models. Chapter 6 of Statistical Models in S eds J. M. Chambers and T. J. Hastie, Wadsworth & Brooks/Cole.
Venables, W. N. and Ripley, B. D. (2002) Modern Applied Statistics with S. New York: Springer (4th ed).