confintr (version 1.0.0)

ci_cor: CI for Correlation Coefficients

Description

This function calculates CIs for a population correlation coefficient. For Pearson correlation, "normal" CIs are available (by stats::cor.test()). Also bootstrap CIs are supported (by default "bca", and the only option for rank correlations).

Usage

ci_cor(
  x,
  y = NULL,
  probs = c(0.025, 0.975),
  method = c("pearson", "kendall", "spearman"),
  type = c("normal", "bootstrap"),
  boot_type = c("bca", "perc", "norm", "basic"),
  R = 9999L,
  seed = NULL,
  ...
)

Value

An object of class "cint" containing these components:

• parameter: Parameter specification.

• interval: CI for the parameter.

• estimate: Parameter estimate.

• probs: Lower and upper probabilities.

• type: Type of interval.

• info: Additional description.

Arguments

x

A numeric vector or a matrix/data.frame with exactly two numeric columns.

y

A numeric vector (only used if x is a vector).

probs

Lower and upper probabilities, by default c(0.025, 0.975).

method

Type of correlation coefficient, one of "pearson" (default), "kendall", or "spearman". For the latter two, only bootstrap CIs are supported. The names can be abbreviated.

type

Type of CI. One of "normal" (the default) or "bootstrap" (the only option for rank-correlations).

boot_type

Type of bootstrap CI ("bca", "perc", "norm", "basic"). Only used for type = "bootstrap".

R

The number of bootstrap resamples. Only used for type = "bootstrap".

seed

An integer random seed. Only used for type = "bootstrap".

...

Further arguments passed to boot::boot().

Examples

ci_cor(iris[1:2])
ci_cor(iris[1:2], type = "bootstrap", R = 999)  # Use larger R
ci_cor(iris[1:2], method = "spearman", type = "bootstrap", R = 999)  # Use larger R