parsec (version 1.2.0)

rank_stability: Rank stability analysis in posetic FOD

Description

The function evaluates average ranks and positions in the ranking of the profiles associated to the selected posets generated by different alpha-cuts in the posetic FOD analysis.

Usage

rank_stability(x, ...)
# S3 method for FODposet
rank_stability(x,
    selection = 1:length(x$covers),
    coverage_probability = 0.9,
    error = 10^(-5), ...
)

Arguments

x

object of class FODposet genrated by the function FOD.

selection

numeric vector or a vector of names to select the cover matrices in the argument x.

coverage_probability

least coverage probability of the rank intervals with extremes lower_ranks and upper_ranks.

error

the "distance"" from uniformity in the sampling distribution of linear extensions used to evaluate the average ranks. See idn for details.

…

any of above.

Value

alpha

vector of the selected alpha values that generate the different posets from which ranks are generated.

average_ranks

data frame of average ranks of the different profiles (by columns) at different alpha values (by rows).

lower_ranks

data frame of the lower bounds of the interval for each profile (by columns) at different alpha values (by rows).

upper_ranks

data frame of the upper bounds of the interval for each profile (by columns) at different alpha values (by rows).

ranking

data frame of positions in the ranking of the profiles (by columns) associated to the selected posets generated by the different alpha values (by rows).

resolution

number of elements of the selected posets generated by the different alpha values.

See Also

FFOD, idn

Examples

# NOT RUN {
v1 <- as.ordered(c("a", "b", "c", "d"))
v2 <- 1:3
prof <- var2prof(varmod = list(v1 = as.ordered(c("a", "b", "c", "d")), v2 = 1:3))
np <- nrow(prof$profiles)

k <- 10 # number of populations
set.seed(0)
populations <- as.data.frame(lapply(1:k, function(x) round(runif(np)*100)))
rownames(populations) <- rownames(prof$profiles)
names(populations) <- paste0("P", 1:k)

x <- FFOD(profiles = prof, distributions = populations)

res <- rank_stability(x)
res
# }