Malgosia Lubczynska
2015-Feb-09 14:28 UTC
[R] NA when trying to calculate AIC value for DLNM
Dear all, I am trying to run a sensitivity analysis for a DLNM combined with a case crossover design and select the best parameters based on AIC values for different model set-ups. model <- glm(mortality ~ cb.temp + ns(soo, 7*7) + dow, family=quasipoisson(), my.data) where cb.temp is the crossbasis matrix for the exposure (temperature) However, the output for the AIC calculations equals NA. Does anyone know how to obtain a correct AIC for a DLNM model? Thank you, Gosia [[alternative HTML version deleted]]
Prof Brian Ripley
2015-Feb-09 17:24 UTC
[R] NA when trying to calculate AIC value for DLNM
On 09/02/2015 14:28, Malgosia Lubczynska wrote:> Dear all, > > I am trying to run a sensitivity analysis for a DLNM combined with a case crossover design and select the best parameters based on AIC values for different model set-ups. > > model <- glm(mortality ~ cb.temp + ns(soo, 7*7) + dow, family=quasipoisson(), my.data) > > where cb.temp is the crossbasis matrix for the exposure (temperature) > > However, the output for the AIC calculations equals NA. > Does anyone know how to obtain a correct AIC for a DLNM model?To have an AIC you need to do maximum-likelihood fitting (and have a likelihood ...), something a quasi-Poisson fit does not give you.> > Thank you, > Gosia > > > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. >-- Brian D. Ripley, ripley at stats.ox.ac.uk Emeritus Professor of Applied Statistics, University of Oxford 1 South Parks Road, Oxford OX1 3TG, UK