step {regr} | R Documentation |
Select a formula-based model
step.regr(object, scope = NULL, expand=FALSE, scale = 0, direction = c("both", "backward", "forward"), trace = FALSE, keep = NULL, steps = 1000, k = 2, ...)
object |
an object representing a model of an appropriate class. This is used as the initial model in the stepwise search. |
scope |
defines the range of models examined in the stepwise search. This should be either a single formula, or a list containing components 'upper' and 'lower', both formulae. See the details for how to specify the formulae and how they are used. |
expand |
logical. If TRUE and |
scale |
used in the definition of the AIC statistic for selecting the models, currently only for 'lm', 'aov' and 'glm' models. The default value, '0', indicates the scale should be estimated: see 'extractAIC'. |
direction |
the mode of stepwise search, can be one of
|
trace |
if positive, information is printed for each step. Larger values may give more detailed information. |
keep |
a filter function whose input is a fitted model object and
the associated 'AIC' statistic, and whose output is
arbitrary. Typically |
steps |
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. |
k |
the multiple of the number of degrees of freedom used for the
penalty. Only |
... |
any additional arguments to |
The set of models searched is determined by the scope
argument.
The right-hand-side of its lower
component must be included in
the model, and right-hand-side of the model is included in the
upper
component. If scope
is a single formula, it specifies
the upper
component, and the lower
model is empty. If scope
is missing, the initial model is used as the upper
model.
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
existing terms.
Missing values lead to a reduced dataset:
step.regr
works on the dataset that includes all variables
appearing in scope
and then drops all lines containing
missing values (by applying na.omit
).
The result is the model re-fitted to the dataset with only
the variables used in the final model.
This may lead to an increased number of rows.
This differs from the behavior of step
of package
stats
.
[from step
of package stats
:]
There is a potential problem in using glm
fits with a variable
scale
, as in that case the deviance is not simply related to the
maximized log-likelihood. The "glm"
method for function
extractAIC
makes the appropriate adjustment for a gaussian
family, but may need to be amended for other cases. (The
binomial
and poisson
families have fixed scale
by default
and do not correspond to a particular maximum-likelihood problem
for variable scale
.)
The selected model is returned, with up to two additional components. There is an
anova |
steps taken in the search, |
keep |
if the |
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 (thus
excluding lm, aov
and survreg
fits, for example).
Werner A. Stahel
step
in package stats
, stepAIC
in package MASS
, add1.regr
, drop1.regr
r.fit <- regr(Fertility ~ ., data = swiss) r.step <- step.regr(r.fit) r.step$anova r.st2 <- step.regr(r.fit, k=8, trace=FALSE) r.st2