A svyplan_overlap from design_overlap() is a numeric vector of
overlap fractions indexed by lag, so x[1] and x[12] are plain
numbers ready for an overlap argument. $ reaches the counts and the
rotation behind them.
Usage
# S3 method for class 'svyplan_overlap'
print(x, ...)
# S3 method for class 'svyplan_overlap'
format(x, ...)
# S3 method for class 'svyplan_overlap'
as.double(x, ...)
# S3 method for class 'svyplan_overlap'
as.list(x, ...)
# S3 method for class 'svyplan_overlap'
x[i]
# S3 method for class 'svyplan_overlap'
Ops(e1, e2)
# S3 method for class 'svyplan_overlap'
Math(x, ...)
# S3 method for class 'svyplan_overlap'
x[[i]]
# S3 method for class 'svyplan_overlap'
x$name
# S3 method for class 'svyplan_overlap'
as.data.frame(
x,
row.names = NULL,
optional = FALSE,
stringsAsFactors = FALSE,
validRN = TRUE,
...
)
# S3 method for class 'svyplan_overlap'
x[i] <- value
# S3 method for class 'svyplan_overlap'
x[[i]] <- valueArguments
- x
A
svyplan_overlapobject.- ...
Additional arguments are not supported and produce an error.
- i
Lag to extract, by position or by its name, so
x[12]andx[["12"]]are both the twelve-occasion overlap. The result is a bare number, carrying neither the class nor the lag as a name.- e1, e2
Operands. Arithmetic and comparison return bare numerics, the counts and the rotation describing the profile as computed and not whatever it was transformed into.
- name
Field to extract:
overlap,shared,n_occasion,rotation,lagorlife.- row.names, optional, stringsAsFactors, validRN
Standard
as.data.frame()arguments.- value
Replacement value. Replacement is refused, an overlap being computed from a rotation rather than assembled.
Value
The method-specific result:
print()Returns
xinvisibly.format()Returns a string.
as.double()Returns the overlap vector.
as.data.frame()Returns one row per lag.
Ops()andMath()Return bare numerics.
Replacement is an error.
Details
The values, the shared counts and the rotation describe one design, so
nothing may move the values while leaving the rest: subsetting and
arithmetic return bare numerics, and replacement is an error. That also
settles pmax() and pmin(), which copy the attributes of their first
argument without dispatching to any method and would otherwise return
something still labelled an overlap whose counts no longer follow from its
values. They assign through [<-, so they are refused here too. Convert
with as.double(x) and they work as usual.
See also
design_overlap(), which builds these objects,
design_rotation() for the pattern behind them,
plot.svyplan_overlap() for the rotation chart, and
print.svyplan_schedule() for the operational schedule a rotation
becomes.
Examples
cps <- design_overlap("4-8-4")
cps
#> Rotation overlap (planning)
#>
#> 16-occasion life, 8 in sample each occasion
#> rotation: 4 in, 8 out, 4 in
#>
#> lag shared overlap
#> 1 6 0.75
#> 2 4 0.5
#> 3 2 0.25
#> 4 0 0
#> 5 0 0
#> 6 0 0
#> 7 0 0
#> 8 0 0
#> 9 1 0.125
#> 10 2 0.25
#> 11 3 0.375
#> 12 4 0.5
#> 13 3 0.375
#> 14 2 0.25
#> 15 1 0.125
# one lag, as a bare number, which is what the overlap arguments take
cps[1]
#> [1] 0.75
cps[["12"]]
#> [1] 0.5
# the fields behind the values
cps$shared
#> [1] 6 4 2 0 0 0 0 0 1 2 3 4 3 2 1
cps$rotation
#> Rotation pattern (planning)
#>
#> 16-occasion life, 8 in sample each occasion
#> rotation: 4 in, 8 out, 4 in
as.data.frame(cps)
#> lag shared n_occasion overlap
#> 1 1 6 8 0.750
#> 2 2 4 8 0.500
#> 3 3 2 8 0.250
#> 4 4 0 8 0.000
#> 5 5 0 8 0.000
#> 6 6 0 8 0.000
#> 7 7 0 8 0.000
#> 8 8 0 8 0.000
#> 9 9 1 8 0.125
#> 10 10 2 8 0.250
#> 11 11 3 8 0.375
#> 12 12 4 8 0.500
#> 13 13 3 8 0.375
#> 14 14 2 8 0.250
#> 15 15 1 8 0.125
# arithmetic drops the class rather than carrying stale counts
as.double(cps)[1:3]
#> [1] 0.75 0.50 0.25