FuzzyQ (version 0.1.0)

fuzzyqCI: Compute Confidence Intervals of Clustering Metrics

Description

Computes confidence intervals of clustering metrics based on the bootstrap replicates produced by fuzzyqBoot.

Usage

fuzzyqCI(fq.bs, fq = NULL, method = "pct", c.level = 0.95)

Arguments

fq.bs

A list returned by FuzzyQ::fuzzyqBoot.

fq

A list of class fuzzyq returned by FuzzyQ::fuzzyq. Required only if method = "bc" or method = "bca".

method

String. Specify the method to compute confidence intervals. Any of the following: "pct" (percentile, the default), "bc" (bias corrected), "bca" (bias corrected and accelerated).

c.level

Number within [0,1]. Specify the confidence interval level. Default is 0.95.

Value

A matrix with upper and lower confidence interval limits of clustering metrics.

Examples

# NOT RUN {
data(antsA)
FQAnts <- fuzzyq(antsA, sorting = TRUE)

# Compute species Commonness Indices of species of 1,000 bootstrap
# replicates:
# }
# NOT RUN {
BS.FQAnts <- fuzzyqBoot (antsA, N = 1e3, level='spp')
# }
# NOT RUN {
# Compute 95 % confidence intervals, percentile method, default values:
# }
# NOT RUN {
BS.sppCI1 <- fuzzyqCI(BS.FQAnts)
# }
# NOT RUN {
# Alternatively, 95 % confidence intervals, bias corrected and accelerated
# method:
# }
# NOT RUN {
BS.sppCI2 <- fuzzyqCI(BS.FQAnts, fq=FQAnts, method = "bca")
# }
# NOT RUN {
# Compute global metrics of 1,000 boostrap replicates:
# }
# NOT RUN {
BS.global <- fuzzyqBoot (antsA, N = 1e3, level='global')
# }
# NOT RUN {
# Compute 95 % confidence intervals, bias corrected and accelerated method:
# }
# NOT RUN {
BS.globalCI <- fuzzyqCI(BS.global, fq=FQAnts, method = "bca")
# }
# NOT RUN {
# }