NEWS

---

sensitivity 1.30.1 (2024-08-28)

• fct johnsonshap() has been corrected and does work now, with bootstrap, argument model=NULL and tell() usage

sensitivity 1.30.0 (2024-01-21)

• Nex fct EPtest
• New fct johnsonshap() (Warning: does not work yet)

sensitivity 1.29.0 (2023-08-31)

• New fct johnson()
• Bugs corrected in testHSIC.R concerning some functions' names that have been renamed

sensitivity 1.28.1 (2023-03-19)

• Bug corrected in fct parameterSets()
• Bugs corrected about all the ggplot.xxx() fcts issues (required by CRAN)
• Correction in the tell() fct of sensiHSIC() for direct call

sensitivity 1.28.0 (2022-09-29)

• New fct correlratio()
• Many bugs corrected in fct sensiHSIC()
• Many additional options added in fct sensiHSIC(), especially HSIC-ANOVA indices
• New fct testHSIC() for independence testing based on HSIC indices
• New benchmark matyas.fun() added in "testmodels.R"

sensitivity 1.27.1 (2022-08-09)

• Adding the topic of 'interpretability of machine learning models' in the description of the package
• emvd() fct and names have been changed to pmvd()
• Several bugs corrected in pme_knn()
• R requirements: Changes from 'class()' to 'inherits()' in several fcts
• Corrections in parameterSets(), see https://github.com/cran/sensitivity/pull/2

sensitivity 1.27.0 (2021-10-20)

• New fct pme_knn() for proportional marginal effects computation
• Addition of an example about visualiz. of perturbed law in PLI() fct

sensitivity 1.26.1 (2021-09-02)

• Simplification in shapleyBlockestimation.R and parameter values changes in shapleyBlockestimation.Rd
• extract and ggplot functionalities added in shapleysobol_knn() fct
• new examples for nice visualizations (via plotCI() of plotrix package) in fcts PLIquantile(), PLIsuperquantile(), PLIquantile_multivar() and PLIsuperquantile_multivar()

sensitivity 1.26.0 (2021-07-09)

• New function shapleysobol_knn() which allows optimized and parallelized computations of Shapley effects/Sobol' indices using the nearest-neighbors method. This fct also replicates the same functionalities than sobolshap_knn() that will be removed in a future package's version
• \donttest examples have been changed in the .Rd of several functions in order to shorten their execution (because the tests are judged too long by CRAN)
• Bug corrected in all PLIxxx() fcts: - conditions (delta != 0) replaced by conditions (delta != mu), mu being the input mean) - pti=phi(lambda[K]) is not recomputed at each bootstrap iteration (causing numerical instabilities)
• Bug corrected in PLIquantile_multivar() and PLIsuperquantile_multivar() fcts (conditions "|" instead of "&")
• Correction in docs of PLIquantile_multivar() and PLIsuperquantile_multivar() to precise that the delta are perturbed means
• In the fct shapleyPermrand(), the test about 'Multiserver queue model' has been suppressed (unsolved bug in Xset)

sensitivity 1.25.0 (2021-03-20)

• New function PLIsuperquantile_multivar()
• Bootstrap added in PLIquantile_multivar()
• Bug corrected in weightTSA() (data structures)
• Change in lmg(): new \code{max.iter} argument to increase the maximum number of iterative optimization step for logistic regression
• Change in emvd(): - new \code{max.iter} argument to increase the maximum number of iterative optimization step for logistic regression - new \code{tol} argument to detect spurious variables - faster implementation - recursive emvd computation can now be partially done in parallel
• Add an example with Gaussian inverse transforms in PLI() fct

sensitivity 1.24.1

• Bug for y format in lmg() and emvd(): allow dataframe

sensitivity 1.24.0 (2021-01-11)

• New function lmg() for LMG sensitivity indices
• New function emvd() for E-MVD sensitivity indices
• Package foreach is necessary (Imports)

sensitivity 1.23.1 (2020-12-08)

• Bug of tests of sensiHSIC()
• Bug in sobolshap_knn() (matrix option in argument U)

sensitivity 1.23.0 (2020-11-28)

• New function qosa() computing quantile-oriented sensitivity analysis via the contrast-function formulation
• Corrections in sensiHSIC() for - taking into account categorical inputs (and the choice of their hyperparameters) - taking into account multiple outputs when some are categorical
• Corrections in sobolshap_knn() for taking into account categorical inputs
• Correction in bootstats() relative to the use of sobolrep() fct

sensitivity 1.22.2 (2020-10-25)

• Correction in PLIsuperquantile() for "bias=TRUE" & "type=VAR" case
• New toy function in testmodels(): Friedman function
• New argument 'varoutput' in delsa() in order to be able to take the output variance as the empirical variance from the sample
• New argument 'randperm' in sobolshap_knn() in order to use or not random permutations to estimate Shapley effects

sensitivity 1.22.1 (2020-08-07)

• Correction in sensiHSIC.R with test.method = "No" in order to not perform the statistical tests
• Correction in nested.cpp
• Change of colors in shapleyBlockEstimation.Rd

sensitivity 1.22.0 (2020-07-18)

• New function shapleyBlockEstimation()
• New functionalities in sensiHSIC(): PCA, kernel for categorical inputs and/or outputs, optimization of permutations numbers, convergence plot for p-values, U-stat estimator for conditional HSIC
• Corrected plot with new fct plotMultOut() (for ubiquitous Sobol' indices) in sobolMultOut() and associated sobolxxx() fcts
• New bootstrap process in PLIquantile() and PLIsuperquantile() via the new argument bootsample

sensitivity 1.21.0 (2020-06-02)

• New function sobolshap_knn() (sensitivity analysis via ranking / nearest neighbours)
• New functionalities in sensiHISIC(): aggregated HSIC

sensitivity 1.20.0 (2020-05-18)

• New functions sobolrep() (Sobol' indices estimation based on replicated orthogonal arrays) and sobolrec() (Recursive estimation of Sobol' indices), which use the new functions addelman_const(), discrepancyCriteria_cplus() and maximin_cplus() (discrepancy and maximin criteria using a C++ implementation)
• bootstrap implemented in sobolrank()
• New function shapleyLinearGaussian() which computes Shapley effects for linear models with Gaussian inputs
• bug corrected in shapleySubsetMc() with the modified line: cost=sum(apply(U,1,function(u) round(Ntot/choose(p,sum(u))/(p-1))))

sensitivity 1.19.0 (2020-04-29)

• New author Sebastien Da veiga
• New function sobolrank()
• Functions src() and pcc() adapted to work with logistic GLM model (via the new argument logistic = TRUE)
• New argument 'title' in ggplot.shapleyPermRand and ggplot.shapleyPermEx
• bug solved in shapleyPermEx() in order to compute the CI for the full first order and independent total Sobol' indices
• New argument 'noise' to consider noisy observations in shapleySubsetMc()

sensitivity 1.18.1 (2020-04-07)

• New argument semi (= TRUE or FALSE) in function pcc() in order to compute semi-partial correlation coefficient (SPCC)
• bug (caused by new R version 4.0.0) solved in: - sobolmartinez.R, base.R, morris.R, morris_oat.R and simplex.R (matrix objects now also inherit from class "array") - sobolMultOut.Rd, testmodels.Rd and delsa.Rd (screen devices x11() should not be used in examples)

sensitivity 1.18.0 (2020-03-13)

• New function: weightTSA.R in order to perform Target Sensitivity Analysis by transforming the output sample using aspecific weight function
• Improvements in sensiHSIC(): 1) for the parameters estimation via the new argument crit.option, 2) for a sequential estimation of the pvalue (independence test based on permutations), 3) new argument target in order to perform Target Sensitivity Analysis
• New argument 'bias' in PLIsuperquantile(): "TRUE" gives the previous computation; "FALSE" corrects this computation (the sample points were not weighted by the likelihood ratio when computing the perturbed superquantile)

sensitivity 1.17.1 (2020-02-14)

• bug (caused by new R version 4.0.0) solved in morris_oat.R (matrix objects now also inherit from class "array")
• exception "multiIndex[inputIndex] == 0 & der" solved in PoincareChaosSqCoef.R
• example with given data in sensiHSIc()

sensitivity 1.17.0 (2019-12-20)

• New functions: squaredIntEstim.R and PoincareChaosSqCoef.R (which allows to compute the squared coefficients in generalized chaos via Poincare differential operators)
• Modification in PoincareOptimal.Rd concerning the returned eigenvalues

sensitivity 1.16.3 (2019-12-05)

• Unresolved matrix/array classes related bug on R4.0.0 version in morris.R, morris_oat.R, soboljansen.R, sobolmartinez.R and sobolroalhs.R Examples that do not work have been put in \donttest{} (these examples work in practice)
• Resolved bug in plot.soboljansen() and plot.sobolmartinez when dealing with matrice-type model outputs
• ylabel changed (from S to PLI) when plotting PLI examples
• legends precised (between full and independent Sobol' indices) when plotting Shapley effects

sensitivity 1.16.2 (2019-10-09)

• New functionalities in sensiHSIC() function (statistical tests of independence)
• Graph legends corrected in sensiHSIC(), sensiFdiv(), sb() and sobolMultOut() functions
• Corrected bug in nodeggplot()

sensitivity 1.16.1 (2019-06-30)

• New function nodeggplot() (in order to ebautify the plots)
• ggplot functions for the following functions: pcc, src, sensiFdiv, sensiHSIC, shapleyPermEx, shapleyPermrand, sobol, sobol2002, sobol2007, sobolEff, soboljansen, sobolmara, sobolmartinez, sobolMultOut, sobolowen, sobolroalhs, sobolroauc, sobolsalt, sobolTIIlo, sobolTIIpf, soboltouati
• Functions sobolroalhs() and sobolSalt(): Modifications of plot functions in order to have "XiXj" instead of "Xij""
• New output of function delsa(): Brute values of derivatives
• Test case corrected in pcc.Rd

sensitivity 1.16.0 (2019-05-25)

• Added functions for calculating Shapley indices from data: shapleySubsetMC()
• Added functions for calculating PLI: PLIsuperquantile() and PLIquantile_multivar()
• In the 3 PLI functions related to quantiles: - new arguments in order to have the PLI in percentage - the result also contains the perturbed quantiles values - bootstrap confidence intervals can be obtained
• Added the toy function heterdisc.fun()

sensitivity 1.15.2 (2018-09-02)

• Added functions for Morris method for multidimensional outputs: morrisMultOut()
• Added the toy function linkletter.fun()
• Change the names of the fcts (dtnorm, ..., dtgumbel, ...) by (dnorm.trunc, ...) (reverse dependency pb with ATmet package). Export now the fcts, documentation entries in truncateddistrib.Rd
• Corrected bug: Replace \dontrun by \donttest in all Rd-files
• Corrected bug: package triangle is suggested and called in PoincareOptimal.Rd
• Function in sobolGPmethods.R and sobolpickfreeze.R have been integrated in sobolGP.R
• Detected bug (not corrected) in tell.sobolGP() Unused argument in km(), passed by eval(), update(), tell.sobolGP() ??) => comment in the example of sobolGP.Rd

sensitivity 1.15.1 (2018-07-28)

• Functions shapleyPermEx() corrected (confidence intervals)

sensitivity 1.15.0 (2017-09-23)

• Function "sobolCert.R" has been (temporarily) suppressed (compilation bug unresolved)
• Function "support.R" has been changed by O. Roustant (new plotting functionalities are available, as the function "scatterplot")
• Resolved bug about "sobolmartinez" method (now suppressed) in the sobolMultOut() function
• sobolMultOut() now returns the functional Sobol' indices via the attributes "Sfct" and "Tfct" of the output object

sensitivity 1.14.0 (2017-02-10)

• Added functions for calculating Shapley indices: shapleyPermEx() and shapleyPermRand()
• Added function for calculating first order Sobol' indices via smoothing technique: sobolSmthSpl()