A Penalized Framework for Distributed Lag Non-Linear Models
Open Access
- 30 January 2017
- journal article
- Published by Oxford University Press (OUP) in Biometrics
- Vol. 73 (3) , 938-948
Abstract
Summary: Distributed lag non-linear models (DLNMs) are a modelling tool for describing potentially non-linear and delayed dependencies. Here, we illustrate an extension of the DLNM framework through the use of penalized splines within generalized additive models (GAM). This extension offers built-in model selection procedures and the possibility of accommodating assumptions on the shape of the lag structure through specific penalties. In addition, this framework includes, as special cases, simpler models previously proposed for linear relationships (DLMs). Alternative versions of penalized DLNMs are compared with each other and with the standard unpenalized version in a simulation study. Results show that this penalized extension to the DLNM class provides greater flexibility and improved inferential properties. The framework exploits recent theoretical developments of GAMs and is implemented using efficient routines within freely available software. Real-data applications are illustrated through two reproducible examples in time series and survival analysis.Keywords
Funding Information
- Medical Research Council (MR/M022625/1)
- London School of Hygiene & Tropical Medicine (105609/Z/14/Z)
- German Research Foundation (GR 3793/1-1)
This publication has 30 references indexed in Scilit:
- Modeling exposure–lag–response associations with distributed lag non‐linear modelsStatistics in Medicine, 2013
- Fast Stable Restricted Maximum Likelihood and Marginal Likelihood Estimation of Semiparametric Generalized Linear ModelsJournal of the Royal Statistical Society Series B: Statistical Methodology, 2010
- Distributed lag non‐linear modelsStatistics in Medicine, 2010
- Flexible modeling of the cumulative effects of time‐dependent exposures on the hazardStatistics in Medicine, 2009
- Latency Models for Analyses of Protracted ExposuresEpidemiology, 2009
- Using tensor product splines in modeling exposure–time–response relationships: Application to the Colorado Plateau Uranium Miners cohortStatistics in Medicine, 2008
- Models for the Relationship Between Ambient Temperature and Daily MortalityEpidemiology, 2006
- Generalized additive distributed lag models: quantifying mortality displacementBiostatistics, 2000
- Models for Exposure-Time-Response Relationships with Applications to Cancer EpidemiologyAnnual Review of Public Health, 1988
- The Distributed Lag Between Capital Appropriations and ExpendituresEconometrica, 1965