confint_betabinom.default: Beta Binomial Confidence Bounds for Quantiles and Probabilities

Description

This non-parametric approach computes the beta binomial bounds (BB) for quantiles and failure probabilities using a procedure similar to the calculation of probabilities in terms of (Median Ranks).

Usage

# S3 method for default
confint_betabinom(
  x,
  status,
  dist_params,
  distribution = c("weibull", "lognormal", "loglogistic", "normal", "logistic", "sev",
    "weibull3", "lognormal3", "loglogistic3"),
  b_lives = c(0.01, 0.1, 0.5),
  bounds = c("two_sided", "lower", "upper"),
  conf_level = 0.95,
  direction = c("y", "x"),
  ...
)

Value

A tibble with class wt_confint containing the following columns:

x : An ordered sequence of the lifetime characteristic regarding the failed units, starting at min(x) and ending up at max(x). With b_lives = c(0.01, 0.1, 0.5) the 1%, 10% and 50% quantiles are additionally included in x, but only if the specified probabilities are in the range of the estimated probabilities.
rank : Interpolated ranks as a function of probabilities, computed with the converted approximation formula of Benard.
prob : An ordered sequence of probabilities with specified b_lives included.
lower_bound : Provided, if bounds is one of "two_sided" or "lower". Lower limit of the confidence region with respect to direction, i.e. quantiles or probabilities.
upper_bound : Provided, if bounds is one of "two_sided" or "upper". Upper limit of the confidence region with respect to direction, i.e. quantiles or probabilities.
distribution : Specified distribution.
bounds : Specified bound(s).
direction : Specified direction.
cdf_estimation_method : A character that is always NA_character. Due to the generic visualization functions this column has to be provided.

Arguments

x: A numeric vector which consists of lifetime data. Lifetime data could be every characteristic influencing the reliability of a product, e.g. operating time (days/months in service), mileage (km, miles), load cycles.
status: A vector of binary data (0 or 1) indicating whether unit i is a right censored observation (= 0) or a failure (= 1).
dist_params: A (named) numeric vector of (log-)location-scale parameters returned from rank_regression.
distribution: Supposed distribution of the random variable. Has to be in line with the specification made in rank_regression.
b_lives: A numeric vector indicating the probabilities p of the \(B_p\)-lives (quantiles) to be considered.
bounds: A character string specifying of which bounds have to be computed. One of "two_sided", "lower" or "upper".
conf_level: Confidence level of the interval.
direction: A character string specifying the direction of the confidence interval. One of "y" (failure probabilities) or "x" (quantiles).
...: Further arguments passed to or from other methods. Currently not used.

Details

The difference to Median Ranks, i.e. finding the probability of rank j at a 50% level, is to determine the probability of rank j on another level, the specified confidence level.

Examples

Run this code

# Vectors:
obs <- seq(10000, 100000, 10000)
status_1 <- c(0, 1, 1, 0, 0, 0, 1, 0, 1, 0)

cycles <- alloy$cycles
status_2 <- alloy$status

# Probability estimation:
prob_tbl <- estimate_cdf(
  x = obs,
  status = status_1,
  method = "johnson"
)

prob_tbl_2 <- estimate_cdf(
  x = cycles,
  status = status_2,
  method = "johnson"
)

# Model estimation with rank_regression():
rr <- rank_regression(
  x = prob_tbl$x,
  y = prob_tbl$prob,
  status = prob_tbl$status,
  distribution = "weibull",
  conf_level = 0.9
)

rr_2 <- rank_regression(
  x = prob_tbl_2$x,
  y = prob_tbl_2$prob,
  status = prob_tbl_2$status,
  distribution = "lognormal3"
)

# Example 1 - Two-sided 95% confidence interval for probabilities ('y'):
conf_betabin_1 <- confint_betabinom(
  x = prob_tbl$x,
  status = prob_tbl$status,
  dist_params = rr$coefficients,
  distribution = "weibull",
  bounds = "two_sided",
  conf_level = 0.95,
  direction = "y"
)

# Example 2 - One-sided lower/upper 90% confidence interval for quantiles ('x'):
conf_betabin_2_1 <- confint_betabinom(
  x = prob_tbl$x,
  status = prob_tbl$status,
  dist_params = rr$coefficients,
  distribution = "weibull",
  bounds = "lower",
  conf_level = 0.9,
  direction = "x"
)

conf_betabin_2_2 <- confint_betabinom(
  x = prob_tbl$x,
  status = prob_tbl$status,
  dist_params = rr$coefficients,
  distribution = "weibull",
  bounds = "upper",
  conf_level = 0.9,
  direction = "x"
)

# Example 3 - Two-sided 90% confidence intervals for both directions using
# a three-parametric model:

conf_betabin_3_1 <- confint_betabinom(
  x = prob_tbl_2$x,
  status = prob_tbl_2$status,
  dist_params = rr_2$coefficients,
  distribution = "lognormal3",
  bounds = "two_sided",
  conf_level = 0.9,
  direction = "y"
)

conf_betabin_3_2 <- confint_betabinom(
  x = prob_tbl_2$x,
  status = prob_tbl_2$status,
  dist_params = rr_2$coefficients,
  distribution = "lognormal3",
  bounds = "two_sided",
  conf_level = 0.9,
  direction = "x"
)