scheidan / adaptmcmc Goto Github PK

I am publishing on CRAN an R-package that calls the MCMC() function during the tests. I got a warning from CRAN that the progress bars should appear only in interactive sessions. Is it possible to hide this progress bar with the current version or is it possible to release a new adaptMCMC version which has a showProgressBar argument in the function MCMC()?

Dear Andreas,

I'd like to cite the package in a publication (and this is quite urgent, because I have to submit the revised article proof today). Calling citation("adaptMCMC") in R returns an error:

citation("adaptMCMC")
Error in if (!is.null(family) && !is.na(match(family, c(jr, tolower(jr))))) { :
missing value where TRUE/FALSE needed

traceback()
11: FUN(X[[i]], ...)
10: lapply(x, as_person1)
9: do.call("c", lapply(x, as_person1))
8: as.list(do.call("c", lapply(x, as_person1)))
7: as.person.default(X[[i]], ...)
6: FUN(X[[i]], ...)
5: lapply(x, as.person)
4: do.call("c", lapply(x, as.person))
3: as.personList.default(meta$Author)
2: as.personList(meta$Author)
1: citation("adaptMCMC")

For my analysis, I've mostly used version 1.1 of the package, so I am citing this version in my article. Is it possible to give the following reference:

Andreas Scheidegger (2012).
adaptMCMC: adaptive Monte Carlo Markov Chain sampler with coerced acceptance rate. R package version 1.1.
https://CRAN.R-project.org/package=adaptMCMC

A BibTeX entry for LaTeX users is

@Manual{,
title = {adaptMCMC: adaptive Monte Carlo Markov Chain sampler with coerced acceptance rate},
author = {Andreas Scheidegger},
year = {2012},
note = {R package version 1.1},
url = {https://CRAN.R-project.org/package=adaptMCMC}
}

Steven Smit:
It’d be a nice (small) feature to allow one to use MCMC.add.samples and change the adaptation settings (for example turning them off).

See https://github.com/helske/ramcmc

gamma == 0.5 should not be allowed

Dear Andreas,

Will it be possible to display a progress bar when using the MCMC parallel. I have an intense function for computing likelihood. It would be nice if progress bar can be shown

Thanks,
Arnab