The regsubsets
function in the leaps package finds
optimal subsets of predictors based on some criterion statistic. This function plots a measure of fit against subset size.
subsets(object, ...)# S3 method for regsubsets
subsets(object,
names=abbreviate(object$xnames, minlength = abbrev),
abbrev=1, min.size=1, max.size=length(names),
legend="interactive",
statistic=c("bic", "cp", "adjr2", "rsq", "rss"),
las=par('las'), cex.subsets=1, ...)
NULL
if the legend
is TRUE
; otherwise a data frame with the legend.
a regsubsets
object produced by the regsubsets
function
in the leaps package.
a vector of (short) names for the predictors, excluding the
regression intercept, if one is present; if missing, these are
derived from the predictor names in object
.
minimum number of characters to use in abbreviating predictor names.
minimum size subset to plot; default is 1
.
maximum size subset to plot; default is number of predictors.
If not FALSE
, in which case the legend is suppressed,
the coordinates at which to place a legend of the abbreviated predictor names on the plot,
in a form recognized by
the legend
function. If "interactive"
, the legend is placed on the
plot interactively with the mouse. By expanding the left or right plot margin, you can place
the legend in the margin, if you wish (see par
).
statistic to plot for each predictor subset; one of:
"bic"
, Bayes Information Criterion;
"cp"
, Mallows's \(C_{p}\);
"adjr2"
, \(R^{2}\) adjusted for degrees of freedom;
"rsq"
, unadjusted \(R^{2}\);
"rss"
, residual sum of squares.
if 0
, ticks labels are drawn parallel to the
axis; set to 1
for horizontal labels (see par
).
can be used to change the relative size of the characters used to
plot the regression subsets; default is 1
.
arguments to be passed down to
subsets.regsubsets
and plot
.
John Fox
Fox, J. (2016) Applied Regression Analysis and Generalized Linear Models, Third Edition. Sage.
Fox, J. and Weisberg, S. (2019) An R Companion to Applied Regression, Third Edition, Sage.
regsubsets
if (require(leaps)){
subsets(regsubsets(undercount ~ ., data=Ericksen),
legend=c(3.5, -37))
}
Run the code above in your browser using DataLab