On 20 October 2022 at 13:16, Elysée Aristide wrote: | Thank you very much for your answers. | | I finally noticed that the issue comes from the numerical optimization, in which I use the class MFuncGrad of the package RcppNumerical. Is there any problem in using this class? This is the only way I know to compute integrals of perform numerical optimizations in Rcpp. The other way might be: construct the function to be minimized in Rcpp and use the function optim in R. But this is not efficient.
That package is also used by others [1] and you should maybe study their code and setup as it seems unlikely to me that your woes are due to RcppNumerical. It likely is a simple inadvertent issue in your package. Dirk [1] https://cloud.r-project.org/web/packages/RcppNumerical/index.html shows Reverse imports: BayesPPD, ccdf, glmdisc, MCPModPack, secr, teachingApps Reverse linking to: apeglm, BayesPPD, BVSNLP, CDatanet, COMIX, fido, gbeta, glmdisc, MCPModPack, OwenQ, secr, teachingApps, vennLasso, ztpln which should give you plenty of examples. -- dirk.eddelbuettel.com | @eddelbuettel | e...@debian.org ______________________________________________ R-package-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-package-devel