onesampb function

One-sample percentile bootstrap

Tests whether a robust location measure (median, Huber Psi) differs from a null value and reports a 95% confidence interval based on percentile bootstrap.

onesampb(x, est = "onestep", nboot = 2000, nv = 0, alpha = 0.05, ...)

Arguments

• x: a numeric vector.
• est: robust estimator to be used ("onestep", "mom", or "median").
• nboot: number of bootstrap samples.
• nv: value under H0.
• alpha: alpha level.
• ...: currently ignored.

Returns

Returns an object of class "onesampb" containing:

• ci: 95% confidence interval

• estimate: robust location sample estimate

• p.value: p-value

• n: number of effective observations

• call: function call

References

Wilcox, R. (2017). Introduction to Robust Estimation and Hypothesis Testing (4th ed.). Elsevier.

See Also

t1way

Examples

set.seed(123)
x <- rnorm(30)
onesampb(x, nboot = 100)    ## H0: Psi = 0

set.seed(123)
x <- rlnorm(30)
onesampb(x, est = "median", nv = 1)   ## H0: median = 1