Stratified Cox / Standard Cox Compound Inference for KK Designs
Source:R/inference_survival_KK_strat_cox.R
InferenceSurvivalKKStratCoxPHIVWC.RdThis class implements a compound estimator for KK matching-on-the-fly designs with survival responses. For matched pairs, it uses stratified Cox proportional hazards regression (each pair is a stratum). For reservoir subjects, it uses standard Cox regression. The two estimates (both log-hazard ratios) are combined via a variance-weighted linear combination.
Computes the randomization distribution of the treatment effect estimate under the sharp null.
Whether compute_rand_two_sided_pval() is actually
usable on this instance right now – FALSE exactly when it
would stop(): an incidence-response instance with no
custom randomization statistic and a design not eligible for
design-randomization-based incidence inference (see
private$should_use_design_randomization_for_incidence()).
TRUE for every other case, including every non-incidence
response type. Public, self-contained (only reads already-set
instance state, no side effects), so InferenceSuite can
check this before attempting the sentinel instead of relying on
the stop() being silently swallowed into a pval = NA
"ok" row – the single source of truth for both this check and
compute_rand_two_sided_pval()'s own guard, so the two can
never drift apart.
Computes a randomization-based p-value.
Value
When debug = FALSE (default), a numeric vector of length r. When
debug = TRUE, a list with: values, errors (list of character
vectors, one per iteration), warnings (list of character vectors, one per
iteration), num_errors, num_warnings,
prop_iterations_with_errors, prop_iterations_with_warnings, and
prop_illegal_values.
A single logical.
Randomization p-value.
Details
Under harden = TRUE, multivariate fits preserve the treatment column and
progressively retry reduced covariate sets after QR-based rank reduction and
correlation-based pruning. Extreme finite coefficients / standard errors are
rejected and treated as non-estimable.
The matched-pair sub-estimate treats each pair as its own stratum (a pair-specific baseline hazard, exactly canceling shared-frailty effects within the pair via Cox's partial likelihood) and is a special case of the Lee-Wei-Amato (1992) large-numbers-of-small-groups stratified Cox approach; the reservoir sub-estimate is a standard unstratified Cox partial-likelihood fit (Cox 1972). The two log-hazard-ratio estimates are combined by inverse-variance weighting, the same rule used throughout the KK IVWC family.
Legacy class. Not fully tested in comprehensive_tests.R.
References
Cox, D. R. (1972). "Regression Models and Life-Tables." Journal of the Royal Statistical Society, Series B, 34(2), 187-220.
Lee, E. W., Wei, L. J., and Amato, D. A. (1992). "Cox-Type Regression Analysis for Large Numbers of Small Groups of Correlated Failure Time Observations." In Survival Analysis: State of the Art, 237-247. Springer. doi:10.1007/978-94-015-7983-4_14
Super class
Inference -> InferenceSurvivalKKStratCoxPHIVWC
Methods
+ inherited public methods from Inference
Inference$capabilities()Inference$compute_asymp_confidence_interval()Inference$compute_asymp_two_sided_pval()Inference$compute_estimate()Inference$compute_exact_confidence_interval()Inference$compute_exact_two_sided_pval_for_treatment_effect()Inference$duplicate()Inference$get_analysis_data()Inference$get_covariates()Inference$get_design_object()Inference$get_model_formula()Inference$get_nonestimable_reason()Inference$get_nonestimable_stage()Inference$get_optimization_alg()Inference$get_response()Inference$get_response_type()Inference$get_treatment()Inference$initialize()Inference$is_nonestimable()Inference$set_optimization_alg()Inference$set_seed()Inference$supports()
InferenceSurvivalKKStratCoxPHIVWC$approximate_randomization_distribution_beta_hat_T()
Usage
InferenceSurvivalKKStratCoxPHIVWC$approximate_randomization_distribution_beta_hat_T(
r = 501,
delta = 0,
transform_responses = "none",
show_progress = TRUE,
permutations = NULL,
debug = FALSE,
zero_one_logit_clamp = .Machine$double.eps
)Arguments
rNumber of randomization vectors. Default 501.
rNumber of randomization vectors.
deltaThe null difference. Default 0.
deltaNull difference.
transform_responsesType of transformation. Default "none".
transform_responsesTransformation.
show_progressShow progress bar. Default TRUE.
show_progressShow progress.
permutationsPre-computed permutations. Default NULL.
permutationsPre-computed permutations.
debugIf
TRUE, return a list with the distribution values and per-iteration diagnostics including error messages, warning messages, counts of each, and summary proportions for iterations with errors, warnings, and illegal (non-finite) values. Runs serially. DefaultFALSE.zero_one_logit_clampThe clamping amount for exact 0 and 1 values when logging
zero_one_logit_clampThe clamping amount for exact 0 and 1 values when logging
InferenceSurvivalKKStratCoxPHIVWC$compute_rand_two_sided_pval()
Usage
InferenceSurvivalKKStratCoxPHIVWC$compute_rand_two_sided_pval(
r = 501,
delta = 0,
transform_responses = "none",
na.rm = TRUE,
show_progress = TRUE,
permutations = NULL,
zero_one_logit_clamp = .Machine$double.eps
)Arguments
rNumber of randomization vectors. Default 501.
rNumber of randomization vectors.
deltaThe null difference. Default 0.
deltaNull difference.
transform_responsesType of transformation. Default "none".
transform_responsesTransformation.
na.rmRemove NAs.
show_progressShow progress bar. Default TRUE.
show_progressShow progress.
permutationsPre-computed permutations. Default NULL.
permutationsPre-computed permutations.
zero_one_logit_clampThe clamping amount for exact 0 and 1 values when logging
zero_one_logit_clampThe clamping amount for exact 0 and 1 values when logging