Package {AdaptHyCensor}

Type:	Package
Title:	Generalized Inference and Data Generation for Adaptive Progressive Hybrid Censoring Schemes
Version:	0.1.0
Description:	Comprehensive computational tools for data generation, statistical inference, and visual diagnostics under Adaptive Type-I and Adaptive Type-II Progressive Hybrid Censoring Schemes. Users can supply custom probability density functions (PDF), cumulative distribution functions (CDF), survival functions, parameter ranges, and progressive schemes for any univariate lifetime distribution. Parameter estimation methods include Maximum Likelihood Estimation (MLE) using multiple optimization algorithms (Newton-Raphson (NR), Broyden-Fletcher-Goldfarb-Shanno (BFGS), BFGS in R (BFGSR), Berndt-Hall-Hall-Hausman (BHHH), Simulated Annealing (SANN), Conjugate Gradients (CG), and Nelder-Mead (NM)), Bayesian estimation via Gibbs and Metropolis-Hastings (M-H) MCMC sampling, Importance Sampling (IS), and Lindley's approximation. Diagnostic tools provide histograms, dot plots, and autocorrelation function (ACF) plots for model validation. Methods are based on Balakrishnan, Cramer, and Kundu (2023, ISBN:978-0-12-398387-9), Ng, Kundu, and Chan (2009, IEEE Transactions on Reliability, 58, 634-642), Lin and Huang (2012, Journal of Statistical Computation and Simulation, 82, 1005-1018), Lindley (1980, Journal of the Royal Statistical Society, Series B, 42, 223-237), and Berndt, Hall, Hall, and Hausman (1974, Annals of Economic and Social Measurement, 3, 653-665).
License:	GPL (≥ 3)
Encoding:	UTF-8
RoxygenNote:	7.3.3
Depends:	R (≥ 4.0.0)
Imports:	stats, graphics
Suggests:	testthat (≥ 3.0.0)
NeedsCompilation:	no
Packaged:	2026-07-25 04:31:32 UTC; shikhar tyagi
Author:	Shikhar Tyagi [aut, cre], Arvind Pandey [aut], Bhupendra Singh [aut], Vrijesh Tripathi [aut]
Maintainer:	Shikhar Tyagi <shikhar1093tyagi@gmail.com>
Repository:	CRAN
Date/Publication:	2026-08-04 14:10:45 UTC

Generic AIC Function for mle_adapt_fit

Description

Generic AIC Function for mle_adapt_fit

Usage

## S3 method for class 'mle_adapt_fit'
AIC(object, ..., k = 2)

Arguments

Value

A numeric value representing AIC.

Bayesian Estimation via Metropolis-Hastings MCMC under Adaptive Type-I PHCS

Description

Performs Bayesian parameter estimation under Adaptive Type-I Progressive Hybrid Censoring using Metropolis-Hastings / Gibbs MCMC sampling.

Usage

bayes_adapt1_phcs(
  data,
  n,
  m,
  T_thresh,
  R,
  pdf,
  cdf = NULL,
  surf = NULL,
  prior = NULL,
  start,
  N_iter = 5000,
  burn_in = 1000,
  proposal_sd = NULL,
  lower = NULL,
  upper = NULL
)

Arguments

Value

A list containing posterior means, medians, modes, standard deviations, HPD credible intervals, and parameter chain samples.

Examples

set.seed(123)
my_cdf <- function(x, par) pexp(x, rate = par[1])
my_pdf <- function(x, par) dexp(x, rate = par[1])
my_surf <- function(x, par) 1 - pexp(x, rate = par[1])
flat_prior <- function(par) ifelse(par[1] > 0, 0, -Inf)
R_plan <- c(rep(1, 8), rep(0, 4))
dat <- r_adapt1_phcs(
  n = 20, m = 12, T_thresh = 2.5,
  R = R_plan, pdf = my_pdf,
  cdf = my_cdf, par = 0.5
)$data
bayes_res <- bayes_adapt1_phcs(
  data = dat, n = 20, m = 12,
  T_thresh = 2.5, R = R_plan,
  pdf = my_pdf, cdf = my_cdf,
  surf = my_surf, prior = flat_prior,
  start = c(rate = 0.8), N_iter = 1000,
  burn_in = 200
)
print(bayes_res$summary)

Bayesian Estimation via Metropolis-Hastings MCMC under Adaptive Type-II PHCS

Description

Performs Bayesian parameter estimation under Adaptive Type-II Progressive Hybrid Censoring using Metropolis-Hastings / Gibbs MCMC sampling.

Usage

bayes_adapt2_phcs(
  data,
  n,
  m,
  T_thresh,
  R,
  pdf,
  cdf = NULL,
  surf = NULL,
  prior = NULL,
  start,
  N_iter = 5000,
  burn_in = 1000,
  proposal_sd = NULL,
  lower = NULL,
  upper = NULL
)

Arguments

Value

A list containing posterior means, medians, modes, standard deviations, HPD credible intervals, and parameter chain samples.

Examples

set.seed(123)
my_cdf <- function(x, par) pexp(x, rate = par[1])
my_pdf <- function(x, par) dexp(x, rate = par[1])
my_surf <- function(x, par) 1 - pexp(x, rate = par[1])
flat_prior <- function(par) ifelse(par[1] > 0, 0, -Inf)
R_plan <- c(rep(1, 8), rep(0, 4))
dat <- r_adapt2_phcs(
  n = 20, m = 12, T_thresh = 2.5,
  R = R_plan, pdf = my_pdf,
  cdf = my_cdf, par = 0.5
)$data
bayes_res <- bayes_adapt2_phcs(
  data = dat, n = 20, m = 12,
  T_thresh = 2.5, R = R_plan,
  pdf = my_pdf, cdf = my_cdf,
  surf = my_surf, prior = flat_prior,
  start = c(rate = 0.8), N_iter = 1000,
  burn_in = 200
)
print(bayes_res$summary)

Importance Sampling Estimation under Adaptive Type-I PHCS

Description

Estimates parameters and posterior expectations via Importance Sampling under Adaptive Type-I PHCS.

Usage

is_adapt1_phcs(
  data,
  n,
  m,
  T_thresh,
  R,
  pdf,
  cdf = NULL,
  surf = NULL,
  prior = NULL,
  start,
  proposal_cov = NULL,
  N_samples = 5000,
  lower = NULL,
  upper = NULL
)

Arguments

Value

A list containing estimated posterior mean, standard error, effective sample size (ESS), and importance weights.

Examples

set.seed(123)
my_cdf <- function(x, par) pexp(x, rate = par[1])
my_pdf <- function(x, par) dexp(x, rate = par[1])
my_surf <- function(x, par) 1 - pexp(x, rate = par[1])
flat_prior <- function(par) ifelse(par[1] > 0, 0, -Inf)
R_plan <- c(rep(1, 8), rep(0, 4))
dat <- r_adapt1_phcs(
  n = 20, m = 12, T_thresh = 2.5,
  R = R_plan, pdf = my_pdf,
  cdf = my_cdf, par = 0.5
)$data
is_res <- is_adapt1_phcs(
  data = dat, n = 20, m = 12,
  T_thresh = 2.5, R = R_plan,
  pdf = my_pdf, cdf = my_cdf,
  surf = my_surf, prior = flat_prior,
  start = c(rate = 0.5), N_samples = 1000
)
print(is_res$summary)

Importance Sampling Estimation under Adaptive Type-II PHCS

Description

Estimates parameters and posterior expectations via Importance Sampling under Adaptive Type-II PHCS.

Usage

is_adapt2_phcs(
  data,
  n,
  m,
  T_thresh,
  R,
  pdf,
  cdf = NULL,
  surf = NULL,
  prior = NULL,
  start,
  proposal_cov = NULL,
  N_samples = 5000,
  lower = NULL,
  upper = NULL
)

Arguments

Value

A list containing estimated posterior mean, standard error, effective sample size (ESS), and importance weights.

Examples

set.seed(123)
my_cdf <- function(x, par) pexp(x, rate = par[1])
my_pdf <- function(x, par) dexp(x, rate = par[1])
my_surf <- function(x, par) 1 - pexp(x, rate = par[1])
flat_prior <- function(par) ifelse(par[1] > 0, 0, -Inf)
R_plan <- c(rep(1, 8), rep(0, 4))
dat <- r_adapt2_phcs(
  n = 20, m = 12, T_thresh = 2.5,
  R = R_plan, pdf = my_pdf,
  cdf = my_cdf, par = 0.5
)$data
is_res <- is_adapt2_phcs(
  data = dat, n = 20, m = 12,
  T_thresh = 2.5, R = R_plan,
  pdf = my_pdf, cdf = my_cdf,
  surf = my_surf, prior = flat_prior,
  start = c(rate = 0.5), N_samples = 1000
)
print(is_res$summary)

Bayesian Estimation via Lindley's Approximation under Adaptive Type-I PHCS

Description

Computes Bayes estimates of parameters using Lindley's expansion for Adaptive Type-I PHCS.

Usage

lindley_adapt1_phcs(
  data,
  n,
  m,
  T_thresh,
  R,
  pdf,
  cdf = NULL,
  surf = NULL,
  prior = NULL,
  start,
  lower = NULL,
  upper = NULL
)

Arguments

Value

A list containing Lindley Bayes estimates, MLEs, and log-likelihood details.

Examples

set.seed(123)
my_cdf <- function(x, par) pexp(x, rate = par[1])
my_pdf <- function(x, par) dexp(x, rate = par[1])
my_surf <- function(x, par) 1 - pexp(x, rate = par[1])
flat_prior <- function(par) ifelse(par[1] > 0, 0, -Inf)
R_plan <- c(rep(1, 8), rep(0, 4))
dat <- r_adapt1_phcs(
  n = 20, m = 12, T_thresh = 2.5,
  R = R_plan, pdf = my_pdf,
  cdf = my_cdf, par = 0.5
)$data
lindley_res <- lindley_adapt1_phcs(
  data = dat, n = 20, m = 12,
  T_thresh = 2.5, R = R_plan,
  pdf = my_pdf, cdf = my_cdf,
  surf = my_surf, prior = flat_prior,
  start = c(rate = 0.8)
)
print(lindley_res$estimates)

Bayesian Estimation via Lindley's Approximation under Adaptive Type-II PHCS

Description

Computes Bayes estimates of parameters using Lindley's expansion for Adaptive Type-II PHCS.

Usage

lindley_adapt2_phcs(
  data,
  n,
  m,
  T_thresh,
  R,
  pdf,
  cdf = NULL,
  surf = NULL,
  prior = NULL,
  start,
  lower = NULL,
  upper = NULL
)

Arguments

Value

A list containing Lindley Bayes estimates, MLEs, and log-likelihood details.

Examples

set.seed(123)
my_cdf <- function(x, par) pexp(x, rate = par[1])
my_pdf <- function(x, par) dexp(x, rate = par[1])
my_surf <- function(x, par) 1 - pexp(x, rate = par[1])
flat_prior <- function(par) ifelse(par[1] > 0, 0, -Inf)
R_plan <- c(rep(1, 8), rep(0, 4))
dat <- r_adapt2_phcs(
  n = 20, m = 12, T_thresh = 2.5,
  R = R_plan, pdf = my_pdf,
  cdf = my_cdf, par = 0.5
)$data
lindley_res <- lindley_adapt2_phcs(
  data = dat, n = 20, m = 12,
  T_thresh = 2.5, R = R_plan,
  pdf = my_pdf, cdf = my_cdf,
  surf = my_surf, prior = flat_prior,
  start = c(rate = 0.8)
)
print(lindley_res$estimates)

Maximum Likelihood Estimation under Adaptive Type-I Progressive Hybrid Censoring

Description

Computes Maximum Likelihood Estimates (MLE) under the Adaptive Type-I Progressive Hybrid Censoring Scheme using various optimization algorithms ("NR", "BFGS", "BFGSR", "BHHH", "SANN", "CG", "NM").

Usage

mle_adapt1_phcs(
  data,
  n,
  m,
  T_thresh,
  R,
  pdf,
  cdf = NULL,
  surf = NULL,
  start,
  method = "NR",
  lower = NULL,
  upper = NULL,
  max_iter = 500,
  ...
)

## S3 method for class 'mle_adapt_fit'
print(x, ...)

## S3 method for class 'mle_adapt_fit'
summary(object, ...)

## S3 method for class 'summary.mle_adapt_fit'
print(x, ...)

## S3 method for class 'mle_adapt_fit'
coef(object, ...)

## S3 method for class 'mle_adapt_fit'
vcov(object, ...)

## S3 method for class 'mle_adapt_fit'
logLik(object, ...)

## S3 method for class 'mle_adapt_fit'
stdEr(object, ...)

Arguments

Value

An S3 object of class mle_adapt_fit containing estimates, standard errors, log-likelihood, Hessian, covariance matrix, and model info.

Methods return printed summary, coefficients, variance-covariance matrix, log-likelihood, standard errors, or AIC.

Examples

set.seed(123)
my_cdf <- function(x, par) pexp(x, rate = par[1])
my_pdf <- function(x, par) dexp(x, rate = par[1])
my_surf <- function(x, par) 1 - pexp(x, rate = par[1])
R_plan <- c(rep(1, 8), rep(0, 4))
dat <- r_adapt1_phcs(
  n = 20, m = 12, T_thresh = 2.5,
  R = R_plan, pdf = my_pdf,
  cdf = my_cdf, par = 0.5
)$data
fit <- mle_adapt1_phcs(
  data = dat, n = 20, m = 12,
  T_thresh = 2.5, R = R_plan,
  pdf = my_pdf, cdf = my_cdf,
  surf = my_surf, start = c(rate = 1.0),
  method = "BFGS"
)
summary(fit)

Maximum Likelihood Estimation under Adaptive Type-II Progressive Hybrid Censoring

Description

Computes Maximum Likelihood Estimates (MLE) under the Adaptive Type-II Progressive Hybrid Censoring Scheme using various optimization algorithms ("NR", "BFGS", "BFGSR", "BHHH", "SANN", "CG", "NM").

Usage

mle_adapt2_phcs(
  data,
  n,
  m,
  T_thresh,
  R,
  pdf,
  cdf = NULL,
  surf = NULL,
  start,
  method = "NR",
  lower = NULL,
  upper = NULL,
  max_iter = 500,
  ...
)

Arguments

Value

An S3 object of class mle_adapt_fit containing estimates, standard errors, log-likelihood, Hessian, covariance matrix, and model info.

Examples

set.seed(123)
my_cdf <- function(x, par) pexp(x, rate = par[1])
my_pdf <- function(x, par) dexp(x, rate = par[1])
my_surf <- function(x, par) 1 - pexp(x, rate = par[1])
R_plan <- c(rep(1, 8), rep(0, 4))
dat <- r_adapt2_phcs(
  n = 20, m = 12, T_thresh = 2.5,
  R = R_plan, pdf = my_pdf,
  cdf = my_cdf, par = 0.5
)$data
fit <- mle_adapt2_phcs(
  data = dat, n = 20, m = 12,
  T_thresh = 2.5, R = R_plan,
  pdf = my_pdf, cdf = my_cdf,
  surf = my_surf, start = c(rate = 1.0),
  method = "BFGS"
)
summary(fit)

Diagnostic Plots for Adaptive Type-I Progressive Hybrid Censoring Scheme

Description

Generates 3-panel diagnostic graphics (Histogram with density overlay, Dot plot, and ACF plot) under Adaptive Type-I PHCS.

Usage

plot_adapt1_phcs(
  input,
  T_thresh = NULL,
  pdf = NULL,
  par = NULL,
  main = "Adaptive Type-I PHCS Diagnostics",
  ...
)

Arguments

Value

Invisibly returns NULL. Displays diagnostic plots.

Examples

set.seed(123)
my_cdf <- function(x, par) pexp(x, rate = par[1])
my_pdf <- function(x, par) dexp(x, rate = par[1])
R_plan <- c(rep(1, 8), rep(0, 4))
dat_gen <- r_adapt1_phcs(
  n = 20, m = 12, T_thresh = 2.5,
  R = R_plan, pdf = my_pdf,
  cdf = my_cdf, par = 0.5
)
plot_adapt1_phcs(
  dat_gen, pdf = my_pdf, par = 0.5
)

Diagnostic Plots for Adaptive Type-II Progressive Hybrid Censoring Scheme

Description

Generates 3-panel diagnostic graphics (Histogram with density overlay, Dot plot, and ACF plot) under Adaptive Type-II PHCS.

Usage

plot_adapt2_phcs(
  input,
  T_thresh = NULL,
  pdf = NULL,
  par = NULL,
  main = "Adaptive Type-II PHCS Diagnostics",
  ...
)

Arguments

Value

Invisibly returns NULL. Displays diagnostic plots.

Examples

set.seed(123)
my_cdf <- function(x, par) pexp(x, rate = par[1])
my_pdf <- function(x, par) dexp(x, rate = par[1])
R_plan <- c(rep(1, 8), rep(0, 4))
dat_gen <- r_adapt2_phcs(
  n = 20, m = 12, T_thresh = 2.5,
  R = R_plan, pdf = my_pdf,
  cdf = my_cdf, par = 0.5
)
plot_adapt2_phcs(
  dat_gen, pdf = my_pdf, par = 0.5
)

Random Data Generation under Adaptive Type-I Progressive Hybrid Censoring Scheme

Description

Generates random censored lifetimes under the Adaptive Type-I Progressive Hybrid Censoring Scheme as described by Lin and Huang (2012).

Usage

r_adapt1_phcs(
  n,
  m,
  T_thresh,
  R,
  pdf = NULL,
  cdf = NULL,
  qdf = NULL,
  par = NULL,
  seed = NULL,
  lower = 1e-07,
  upper = 1e+05
)

Arguments

Value

A list containing:

References

Lin, C. T., & Huang, Y. L. (2012). On adaptive Type-I progressive hybrid censored samples. Journal of Statistical Computation and Simulation, 82(7), 1005-1018.

Balakrishnan, N., Cramer, E., & Kundu, D. (2023). Hybrid censoring know-how: Designs and implementations. Academic Press.

Examples

# Example: Standard Exponential distribution
set.seed(123)
my_cdf <- function(x, par) pexp(x, rate = par[1])
my_pdf <- function(x, par) dexp(x, rate = par[1])
my_qdf <- function(p, par) qexp(p, rate = par[1])
R_plan <- c(rep(1, 8), rep(0, 4))
res <- r_adapt1_phcs(
  n = 20, m = 12, T_thresh = 2.5,
  R = R_plan, pdf = my_pdf,
  cdf = my_cdf, qdf = my_qdf, par = 0.5
)
print(res$data)

Random Data Generation under Adaptive Type-II Progressive Hybrid Censoring Scheme

Description

Generates random censored lifetimes under the Adaptive Type-II Progressive Hybrid Censoring Scheme as described by Ng, Kundu, and Chan (2009).

Usage

r_adapt2_phcs(
  n,
  m,
  T_thresh,
  R,
  pdf = NULL,
  cdf = NULL,
  qdf = NULL,
  par = NULL,
  seed = NULL,
  lower = 1e-07,
  upper = 1e+05
)

Arguments

Value

A list containing:

References

Ng, H. K. T., Kundu, D., & Chan, P. S. (2009). Statistical inference for a progressive line-testing experiment with adaptive progressive Type-II censoring. IEEE Transactions on Reliability, 58(4), 634-642.

Balakrishnan, N., Cramer, E., & Kundu, D. (2023). Hybrid censoring know-how: Designs and implementations. Academic Press.

Examples

# Example: Standard Exponential distribution
set.seed(123)
my_cdf <- function(x, par) pexp(x, rate = par[1])
my_pdf <- function(x, par) dexp(x, rate = par[1])
my_qdf <- function(p, par) qexp(p, rate = par[1])
R_plan <- c(rep(1, 8), rep(0, 4))
res <- r_adapt2_phcs(
  n = 20, m = 12, T_thresh = 2.5,
  R = R_plan, pdf = my_pdf,
  cdf = my_cdf, qdf = my_qdf, par = 0.5
)
print(res$data)

Generic Standard Error Function

Description

Generic Standard Error Function

Usage

stdEr(object, ...)

Arguments

Value

A numeric vector of standard errors.