The bbqr package as a whole is distributed under GPL (>= 2).

Third-party components
----------------------

src/wrapper.c

    The four-line bridge exposing R's RNG (GetRNGstate, PutRNGstate,
    unif_rand, norm_rand) to Fortran is taken from the bayesQR package by
    Dries F. Benoit, Rahim Al-Hamzawi, Keming Yu and Dirk Van den Poel,
    distributed on CRAN under GPL (>= 2). The general layout of the package,
    including the convention of handling the intercept separately from the
    penalised slopes, also follows bayesQR.

    bayesQR is described in:
      Benoit, D. F. and Van den Poel, D. (2017). bayesQR: A Bayesian Approach
      to Quantile Regression. Journal of Statistical Software, 76(7), 1-32.
      doi:10.18637/jss.v076.i07

    These authors are credited in the Authors@R field of DESCRIPTION with the
    roles "ctb" (contributor) and, for Dries F. Benoit, "cph" (copyright
    holder).

Original components
-------------------

All six MCMC kernels shipped in this package were written for it.

Binary response, observed through a threshold:

    src/QRb_BQR_mcmc.f95   unpenalised binary quantile regression,
                           implementing Benoit and Van den Poel (2012)
    src/QRb_L_mcmc.f95     lasso binary quantile regression,
                           implementing Benoit, Al-Hamzawi and Yu (2013)
    src/QRb_AL_mcmc.f95    adaptive-lasso binary quantile regression, with
                           the identification anchors described in ?bbqr

Observed continuous response:

    src/QRc_BQR_mcmc.f95   unpenalised quantile regression; the kernel is the
                           sampler of Kozumi and Kobayashi (2011)
    src/QRc_L_mcmc.f95     lasso quantile regression, carrying the hierarchy
                           of Benoit, Al-Hamzawi and Yu (2013) over to an
                           observed response
    src/QRc_AL_mcmc.f95    adaptive-lasso quantile regression; at its default
                           q = 1 it reproduces the penalty of Alhamzawi, Yu
                           and Benoit (2012)

They implement published algorithms, which are cited on the corresponding
help pages and in inst/CITATION, but the code is not derived from the bayesQR
kernels.