Bandwidth for density() via Normal Reference Distribution

A well-supported rule-of-thumb for choosing the bandwidth of a Gaussian kernel density estimator.

bandwidth.nrd(x)

Arguments

x	A data vector.

Value

A bandwidth on a scale suitable for the width argument of density.

References

Venables, W. N. and Ripley, B. D. (2002) Modern Applied Statistics with S. Springer, equation (5.5) on page 130.

Examples

# The function is currently defined as
function(x)
{
    r <- quantile(x, c(0.25, 0.75))
    h <- (r[2] - r[1])/1.34
    4 * 1.06 * min(sqrt(var(x)), h) * length(x)^(-1/5)
}
#> function(x)
#> {
#>     r <- quantile(x, c(0.25, 0.75))
#>     h <- (r[2] - r[1])/1.34
#>     4 * 1.06 * min(sqrt(var(x)), h) * length(x)^(-1/5)
#> }
#> <environment: 0xe5f4f20>