groupMeans( x, group=NULL, na.rm=F, weights=NULL, repeats=F, ...) groupSums( x, group=NULL, na.rm=F, weights=NULL, repeats=F, ...) groupVars( x, group=NULL, na.rm=F, weights=NULL, repeats=F, ...) groupStdevs(x, group=NULL, na.rm=F, weights=NULL, repeats=F, ...) groupMaxs(x, group=NULL, na.rm=F, weights=NULL, repeats=F, ...) groupMins(x, group=NULL, na.rm=F, weights=NULL, repeats=F, ...) groupProds(x, group=NULL, na.rm=F, weights=NULL, repeats=F, ...) groupRanges(x, group=NULL, na.rm=F, weights=NULL, repeats=F, ...) groupAnys(x, group=NULL, na.rm=F, weights=NULL, repeats=F, ...) groupAlls(x, group=NULL, na.rm=F, weights=NULL, repeats=F, ...) subtractMeans(x, group=NULL, na.rm=F, weights=NULL)
numRows(x). Unique values of
this vector determine groups (sets of rows).
Statistics are
calculated for each group separately, and for each column.
The default corresponds to a single group.
FALSE, missing values (
NA) in the input result in
missing values in corresponding elements of the output.
If
TRUE then missing values are omitted from calculations.
numRows(x). If present then weighted
statistics are calculated.
T,
F,
0,
1, or
2. See "VALUE" below.
groupVars and
groupStdevs accept arguments
unbiased and
SumSquares;
see
.
repeats=F or
0 then the
"group" functions return one
statistic value per stratum per column of
x . Thus if
x is an
array, the first dimension of the result is equal to the number of
groups, and all other dimensions are the same as for
x. If
x is a
vector, the functions return a vector (if more than one group) or
scalar (if one group). The names or row names of the result
correspond to the sorted unique values of
group.
The exception is
groupRanges, which returns two values per stratum.
In this case the first dimension of the output corresponds to the two
values, the second dimension to groups, and other dimensions as for
x.
If
repeats=T
or
1 the
"group" functions return an object with the same
dimensions as
x: each entry is the value of the statistic for the
stratum containing the corresponding entry from
x.
(For
groupRanges a dimension of 2 is prepended.)
If
repeats=2,
return a list with both the repeated and non-repeated objects.
The list has names that match the statistic --
"means" and
"repeat.means", for example.
subtractMeans returns a vector or array the same size as
x, with
value equal to
x minus
"repeat.means".
# Center Mileage data about the averages for each type of vehicle
subtractMeans(fuel.frame$Mileage,group=fuel.frame$Type)
# same, randomly weighted means
subtractMeans(fuel.frame$Mileage,group=fuel.frame$Type, weights=runif(60))
# means and standard deviations for Types Small and Sporty
groupMeans(fuel.frame$Mileage, group=fuel.frame$Type)[c("Small", "Sporty")]
groupStdevs(fuel.frame$Mileage, group=fuel.frame$Type)[c("Small", "Sporty")]