top of page


Welcome to the blapsr project!


Gressani Oswaldo 30/03/2022


The blapsr package can be used for approximate Bayesian inference in survival models and generalized additive models. The theoretical framework for the methodology underlying the routines has been developed in [1] Gressani & Lambert (2018), [2] Gressani & Lambert (2020) and [3] Gressani & Lambert (2021) during my PhD thesis at the University of Louvain. It allows to fit Cox proportional hazards models and promotion time cure models (for right censored data), additive partial linear models with normal errors and generalized additive models for a response belonging to the one-parameter exponential family. Up to date the package is still in its infancy and will (hopefully!) evolve in the future. The aim of this website is to serve as a centralized platform for the blapsr software with dedicated group discussions, announcements relative to package updates, comments, examples and more. 




[1] Gressani, O. and Lambert, P. (2018).  Fast Bayesian inference using Laplace approximations in a flexible promotion time cure model based on P-splines. Computational Statistics and Data Analysis, 124, 151-167.

[2] Gressani, O. and Lambert, P. (2020) The Laplace-P-spline methodology for fast approximate Bayesian inference in additive partial linear models. ISBA Discussion papers.

[3] Gressani, O. and Lambert, P. (2021). Laplace approximations for fast Bayesian inference in generalized additive models based on P-splines. Computational Statistics and Data Analysis, 154.



The Comprehensive R Archive Network (CRAN) hosts the package as from July 13th 2020 with initial version 0.5.1. Metadata on the package, license, copyright and citation information can be found at


To work with the blapsr package from CRAN, simply install the package using install.packages("blapsr") and load it in your library with library("blapsr").




The Github repository  hosts the in-development version of the blapsr package. Currently

the version is Whenever the development version evolves, the fourth component will be incremented, for instance version is a successor of, say, More important updates will be submitted to CRAN and will affect the other three components

of the version.


To work with the  blapsr package from Github, simply install the devtools package using install.packages("devtools"), then type the following command devtools::install_github("oswaldogressani/blapsr")and load it in the library. That's it, enjoy!


A quick dive into the main routines

(Last updated June 30th 2020)

Fit an additive partial linear model with normal errors.

Fit a Cox proportional hazards model for right censored data.

Fit a generalized additive model.

Fit a promotion time cure model.

bottom of page