Query Composite Hypotheses

---

Documentation for package ‘qch’ version 2.1.3

• DESCRIPTION file.

Help Pages

Copula.Hconfig_gaussian_density	Gaussian copula density for each H-configuration.
EM_calibration_gaussian	EM calibration in the case of the Gaussian copula (unsigned)
EM_calibration_gaussian_memory	EM calibration in the case of the Gaussian copula (unsigned) with memory management
EM_calibration_indep	EM calibration in the case of conditional independence
EM_calibration_indep_memory	EM calibration in the case of conditional independence with memory management (unsigned)
f1_separation_signed	Signed case function: Separate f1 into f+ and f-
FastKerFdr_signed	FastKerFdr signed
FastKerFdr_unsigned	FastKerFdr unsigned
fHconfig_sum_update_gaussian_copula_ptr_parallel	Computation of the sum sum_c(w_c*psi_c) using Gaussian copula parallelized version
fHconfig_sum_update_ptr_parallel	Computation of the sum sum_c(w_c*psi_c) parallelized version
gaussian_copula_density	Gaussian copula density
GetH1AtLeast	Specify the configurations corresponding to the composite H_1 test "AtLeast".
GetH1Equal	Specify the configurations corresponding to the composite H_1 test "Equal".
GetHconfig	Generate the H_0/H_1 configurations.
last_incomplete_trapezoid_arma	This function is a re-implementation of the initial R loop computing last incomplete trapezoid. See R function integral.kde_adapted().
prior_update_arma_ptr_parallel	Update of the prior estimate in EM algo parallelized version
prior_update_gaussian_copula_ptr_parallel	Update of the prior estimate in EM algo using Gaussian copula, parallelized version
PvalSets	Synthetic example to illustrate the main qch functions
PvalSets_cor	Synthetic example to illustrate the main qch functions using Gaussian copula
qch.fit	Infer posterior probabilities of H_0/H_1 configurations.
qch.test	Perform composite hypothesis testing.
R.MLE	Gaussian copula correlation matrix Maximum Likelihood estimator.
R.MLE.check	Check the Gaussian copula correlation matrix Maximum Likelihood estimator
R.MLE.memory	Gaussian copula correlation matrix Maximum Likelihood estimator (memory handling)
remove_decreasing_values_cpp	Same as function above but does not handle the index ordering of the vector q. Therefore, the 2nd argument order_q has to be an index ordered version of the vector q. Indeed, the R base function: order() is twice as fast as the arma::sort_index(q) This is therefore the recommended function to use.
remove_decreasing_values_cpp_slow_ordering	This function is a re-implementation of the initial R side while loop. See the end of R function integral.kde_adapted(). As shown in the commentary below, it is twice as slow to handle the index ordering of the vector q (2nd argument) here with the function arma::sort_index(). Consequently, it is recommended to use the function remove_decreasing_values_cpp() instead.
R_MLE_update_gaussian_copula_ptr_parallel	Update the estimate of R correlation matrix of the gaussian copula, parallelized version