Naive kernel mean-residual-life estimator (4.2)

The baseline whose boundary failure Ch. 4 exists to fix: bias is O(h) near the edge of the support and can degrade to O(1). Compare with morie_fauzi_mrl_boundary_free_2() on the same data.

Usage

morie_fauzi_mrl_naive(x, t_grid, h = NULL)

Arguments

x

numeric sample.

t_grid

evaluation points.

h

bandwidth; 4^(1/3) sigma n^{-1/3} if NULL – the mean residual life is distribution-function-type, not density-type (its Theorem 4.3 variance is O(1/n) - O(h/n)).

Value

list: t_grid, mrl, bandwidth, interior_bias_order, boundary_bias_order, boundary_safe (FALSE), n, method.

References

Fauzi and Maesono (2023), Eq. (4.2).

Examples

set.seed(1)
morie_fauzi_mrl_naive(stats::rexp(200), c(0.5, 1))$mrl
#> [1] 0.8982861 0.8300170