Penalized distributed lag linear and non-linear models

Antonio Gasparrini Affiliation: London School of Hygiene & Tropical Medicine, UK
dlnm version 2.4.10 , 2026-11-21
11footnotetext: This document is included as a vignette (a document created using the R function Sweave()) of the package dlnm. It is automatically downloaded together with the package and can be simply accessed through R by typing vignette("dlnmPenalized").

1 Preamble

This vignette dlnmPenalized illustrates the extension of the R package dlnm to perform a penalized versions of distributed lag linear (DLMs) and non-linear models (DLNMs). This development is thoroughly described in Gasparrini et al. (2017).

The extension of the DLM/DLNM framework to penalized splines within generalized additive models is implemented by embedding the dlnm and mgcv packages. The latter is used primarily as a computational engine for the estimation of smoothed exposure-lag-response relationships, and to a some extent as a tool for deriving the parameterization of the basis functions and penalty terms. Specifically, two approaches to penalization are implemented in dlnm and described here.

A general overview of functions included in the package, with information on its installation and a brief summary of the DLNM methodology are included in the vignette dlnmOverview, which represents the main documentation of dlnm. The user can refer to that vignette for a general introduction to the package.

Please send comments or suggestions and report bugs to antonio.gasparrini@lshtm.ac.uk.

2 Penalized DLMs and DLNMs

This vignette is under development. For an illustration, refer to Gasparrini et al. (2017) and the R code included as supplementary material that reproduces the examples in the article.

References

  • Gasparrini et al. (2017) A. Gasparrini, F. Scheipl, B. Armstrong, and M. G. Kenward A penalized framework for distributed lag non-linear models. Biometrics (In Press). Cited by: §1, §2.